{
 "provenance": {
  "stamped_at": "2026-11-11T13:26:04+00:00",
  "git_sha": "743eee8e",
  "n_models": 1483,
  "n_benchmarks": 1550
 },
 "anchors": {
  "low": {
   "name": "Claude 3.5 Sonnet",
   "sci": 130
  },
  "high": {
   "name": "GPT-5 (high)",
   "sci": 150
  }
 },
 "fit": {
  "n_benchmarks": 908
 },
 "rules": {
  "min_ranked_hard_obs": 5,
  "min_headline_hard_obs": 10
 },
 "corpus": {
  "n_scores": 64602,
  "n_sources": 117
 },
 "counts": {
  "total": 790,
  "shown": 352,
  "hidden": 438,
  "skus": 1120
 },
 "models": [
  {
   "key": "claude fable 52",
   "name": "Claude Fable 5.2",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 167.5,
   "n_obs": 135,
   "n_configs": 2,
   "configs": [
    {
     "name": "Claude Fable 5.2 (max)",
     "sci": 167.5,
     "n": 29,
     "headline": true
    },
    {
     "name": "Claude Fable 5.2 (standard)",
     "sci": 162.5,
     "n": 106
    }
   ],
   "release_date": "2026-11-09",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 1
  },
  {
   "key": "claude mythos 52",
   "name": "Claude Mythos 5.2",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 166.6,
   "n_obs": 14,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-11-09",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "lab-reported only"
   ],
   "rank": 2
  },
  {
   "key": "gemini 4 carbon",
   "name": "Gemini 4 Carbon",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 165.5,
   "n_obs": 107,
   "n_configs": 3,
   "configs": [
    {
     "name": "Gemini 4 Carbon (high)",
     "sci": 165.5,
     "n": 26,
     "headline": true
    },
    {
     "name": "Gemini 4 Carbon (medium)",
     "sci": 163.9,
     "n": 5
    },
    {
     "name": "Gemini 4 Carbon (standard)",
     "sci": 162.5,
     "n": 76
    }
   ],
   "release_date": "2026-10-28",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 3
  },
  {
   "key": "gpt 6 astra",
   "name": "GPT-6 Astra",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 165.0,
   "n_obs": 360,
   "n_configs": 7,
   "configs": [
    {
     "name": "GPT-6 Astra (max)",
     "sci": 165.0,
     "n": 37,
     "headline": true
    },
    {
     "name": "GPT-6 Astra (unknown)",
     "sci": 164.5,
     "n": 6
    },
    {
     "name": "GPT-6 Astra (xhigh)",
     "sci": 163.5,
     "n": 17
    },
    {
     "name": "GPT-6 Astra (medium)",
     "sci": 162.3,
     "n": 15
    },
    {
     "name": "GPT-6 Astra (high)",
     "sci": 162.1,
     "n": 20
    },
    {
     "name": "GPT-6 Astra (standard)",
     "sci": 162.0,
     "n": 247
    },
    {
     "name": "GPT-6 Astra (low)",
     "sci": 161.8,
     "n": 18
    }
   ],
   "release_date": "2026-09-03",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 4
  },
  {
   "key": "claude opus 55",
   "name": "Claude Opus 5.5",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 164.9,
   "n_obs": 328,
   "n_configs": 6,
   "configs": [
    {
     "name": "Claude Opus 5.5 (max)",
     "sci": 164.9,
     "n": 65,
     "headline": true
    },
    {
     "name": "Claude Opus 5.5 (xhigh)",
     "sci": 163.6,
     "n": 9
    },
    {
     "name": "Claude Opus 5.5 (high)",
     "sci": 162.7,
     "n": 16
    },
    {
     "name": "Claude Opus 5.5 (standard)",
     "sci": 162.0,
     "n": 220
    },
    {
     "name": "Claude Opus 5.5 (medium)",
     "sci": 161.8,
     "n": 11
    },
    {
     "name": "Claude Opus 5.5 (low)",
     "sci": 158.2,
     "n": 7
    }
   ],
   "release_date": "2026-09-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 5
  },
  {
   "key": "claude mythos preview",
   "name": "Claude Mythos Preview",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 164.4,
   "n_obs": 15,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-04-07",
   "cost_usd": null,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 6
  },
  {
   "key": "gpt 61 terra",
   "name": "GPT-6.1 Terra",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 164.1,
   "n_obs": 133,
   "n_configs": 6,
   "configs": [
    {
     "name": "GPT-6.1 Terra (max)",
     "sci": 164.1,
     "n": 17,
     "headline": true
    },
    {
     "name": "GPT-6.1 Terra (xhigh)",
     "sci": 161.4,
     "n": 9
    },
    {
     "name": "GPT-6.1 Terra (high)",
     "sci": 160.3,
     "n": 7
    },
    {
     "name": "GPT-6.1 Terra (standard)",
     "sci": 159.5,
     "n": 86
    },
    {
     "name": "GPT-6.1 Terra (medium)",
     "sci": 159.0,
     "n": 7
    },
    {
     "name": "GPT-6.1 Terra (low)",
     "sci": 157.3,
     "n": 7
    }
   ],
   "release_date": "2026-10-26",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 7
  },
  {
   "key": "gpt 61 sol",
   "name": "GPT-6.1 Sol",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 164.0,
   "n_obs": 263,
   "n_configs": 6,
   "configs": [
    {
     "name": "GPT-6.1 Sol (max)",
     "sci": 164.0,
     "n": 24,
     "headline": true
    },
    {
     "name": "GPT-6.1 Sol (xhigh)",
     "sci": 163.1,
     "n": 9
    },
    {
     "name": "GPT-6.1 Sol (high)",
     "sci": 162.2,
     "n": 7
    },
    {
     "name": "GPT-6.1 Sol (medium)",
     "sci": 161.5,
     "n": 8
    },
    {
     "name": "GPT-6.1 Sol (standard)",
     "sci": 161.4,
     "n": 208
    },
    {
     "name": "GPT-6.1 Sol (low)",
     "sci": 159.4,
     "n": 7
    }
   ],
   "release_date": "2026-09-29",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 8
  },
  {
   "key": "claude sonnet 55",
   "name": "Claude Sonnet 5.5",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 163.9,
   "n_obs": 280,
   "n_configs": 6,
   "configs": [
    {
     "name": "Claude Sonnet 5.5 (max)",
     "sci": 163.9,
     "n": 56,
     "headline": true
    },
    {
     "name": "Claude Sonnet 5.5 (xhigh)",
     "sci": 161.4,
     "n": 11
    },
    {
     "name": "Claude Sonnet 5.5 (standard)",
     "sci": 161.3,
     "n": 193
    },
    {
     "name": "Claude Sonnet 5.5 (high)",
     "sci": 158.9,
     "n": 7
    },
    {
     "name": "Claude Sonnet 5.5 (medium)",
     "sci": 154.0,
     "n": 7
    },
    {
     "name": "Claude Sonnet 5.5 (low)",
     "sci": 152.3,
     "n": 6
    }
   ],
   "release_date": "2026-09-28",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 9
  },
  {
   "key": "claude fable 51",
   "name": "Claude Fable 5.1",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 163.2,
   "n_obs": 275,
   "n_configs": 12,
   "configs": [
    {
     "name": "Claude Fable 5.1 (unknown effort)",
     "sci": 165.1,
     "n": 5,
     "thin": true
    },
    {
     "name": "Claude Fable 5.1 (max)",
     "sci": 163.2,
     "n": 28,
     "headline": true
    },
    {
     "name": "Claude Fable 5.1 (high)",
     "sci": 162.9,
     "n": 12
    },
    {
     "name": "Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort)",
     "sci": 162.6,
     "n": 5
    },
    {
     "name": "Claude Fable 5.1 (Adaptive Reasoning, Max Effort)",
     "sci": 162.4,
     "n": 5
    },
    {
     "name": "Claude Fable 5.1 (xhigh)",
     "sci": 162.2,
     "n": 7
    },
    {
     "name": "Claude Fable 5.1 (medium)",
     "sci": 162.2,
     "n": 7
    },
    {
     "name": "Claude Fable 5.1 (Adaptive Reasoning, High Effort)",
     "sci": 162.0,
     "n": 5
    }
   ],
   "release_date": "2026-09-01",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 10
  },
  {
   "key": "claude mythos 51",
   "name": "Claude Mythos 5.1",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 163.1,
   "n_obs": 28,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-01",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 11
  },
  {
   "key": "gemini 4 argon",
   "name": "Gemini 4 Argon",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 163.1,
   "n_obs": 248,
   "n_configs": 4,
   "configs": [
    {
     "name": "Gemini 4 Argon (high)",
     "sci": 163.1,
     "n": 37,
     "headline": true
    },
    {
     "name": "Gemini 4 Argon (medium)",
     "sci": 162.3,
     "n": 5
    },
    {
     "name": "Gemini 4 Argon (standard)",
     "sci": 161.7,
     "n": 206
    }
   ],
   "release_date": "2026-10-13",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 12
  },
  {
   "key": "claude mythos 5",
   "name": "Claude Mythos 5",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 163.0,
   "n_obs": 12,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-06-09",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 13
  },
  {
   "key": "claude fable 5",
   "name": "Claude Fable 5",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 162.9,
   "n_obs": 244,
   "n_configs": 7,
   "configs": [
    {
     "name": "Claude Fable 5 (effort unspecified)",
     "sci": 165.5,
     "n": 7,
     "thin": true
    },
    {
     "name": "Claude Fable 5 (high)",
     "sci": 162.9,
     "n": 15,
     "headline": true
    },
    {
     "name": "Claude Fable 5 (max)",
     "sci": 160.8,
     "n": 33
    },
    {
     "name": "Claude Fable 5 (medium)",
     "sci": 160.5,
     "n": 5
    },
    {
     "name": "Claude Fable 5 (standard)",
     "sci": 159.9,
     "n": 157
    },
    {
     "name": "Claude Fable 5 (xhigh)",
     "sci": 159.6,
     "n": 21
    },
    {
     "name": "Claude Fable 5 (low)",
     "sci": 158.3,
     "n": 6
    }
   ],
   "release_date": "2026-06-09",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 14
  },
  {
   "key": "fugu ultra v2",
   "name": "Fugu Ultra v2",
   "organization": "Sakana AI",
   "country": "Japan",
   "reasoning_status": "reasoning",
   "sci": 162.2,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-11",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "lab-reported only",
    "thin evidence"
   ],
   "rank": 15
  },
  {
   "key": "gpt 6 sol",
   "name": "GPT-6 Sol",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 162.1,
   "n_obs": 208,
   "n_configs": 6,
   "configs": [
    {
     "name": "GPT-6 Sol (max)",
     "sci": 162.1,
     "n": 27,
     "headline": true
    },
    {
     "name": "GPT-6 Sol (xhigh)",
     "sci": 159.7,
     "n": 10
    },
    {
     "name": "GPT-6 Sol (standard)",
     "sci": 159.3,
     "n": 145
    },
    {
     "name": "GPT-6 Sol (high)",
     "sci": 158.4,
     "n": 10
    },
    {
     "name": "GPT-6 Sol (medium)",
     "sci": 156.9,
     "n": 8
    },
    {
     "name": "GPT-6 Sol (low)",
     "sci": 153.2,
     "n": 8
    }
   ],
   "release_date": "2026-09-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 16
  },
  {
   "key": "gpt 55 pro",
   "name": "GPT-5.5 Pro",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 161.9,
   "n_obs": 15,
   "n_configs": 2,
   "configs": [
    {
     "name": "GPT-5.5 Pro (xhigh)",
     "sci": 161.9,
     "n": 7,
     "headline": true
    },
    {
     "name": "GPT-5.5 Pro (standard)",
     "sci": 160.5,
     "n": 8
    }
   ],
   "release_date": "2026-04-23",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 17
  },
  {
   "key": "claude opus 5",
   "name": "Claude Opus 5",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 161.9,
   "n_obs": 300,
   "n_configs": 11,
   "configs": [
    {
     "name": "Claude Opus 5 (max)",
     "sci": 161.9,
     "n": 52,
     "headline": true
    },
    {
     "name": "Claude Opus 5 (Adaptive Reasoning, Max Effort)",
     "sci": 161.5,
     "n": 5
    },
    {
     "name": "Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)",
     "sci": 161.3,
     "n": 5
    },
    {
     "name": "Claude Opus 5 (Adaptive Reasoning, High Effort)",
     "sci": 161.2,
     "n": 5
    },
    {
     "name": "Claude Opus 5 (xhigh)",
     "sci": 161.1,
     "n": 12
    },
    {
     "name": "Claude Opus 5 (high)",
     "sci": 160.4,
     "n": 17
    },
    {
     "name": "Claude Opus 5 (standard)",
     "sci": 160.4,
     "n": 170
    },
    {
     "name": "Claude Opus 5 (Adaptive Reasoning, Medium Effort)",
     "sci": 159.6,
     "n": 5
    }
   ],
   "release_date": "2026-07-24",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 18
  },
  {
   "key": "kimi k31 pro",
   "name": "Kimi K3.1 Pro",
   "organization": "Moonshot AI",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 161.8,
   "n_obs": 140,
   "n_configs": 2,
   "configs": [
    {
     "name": "Kimi K3.1 Pro (max)",
     "sci": 161.8,
     "n": 29,
     "headline": true,
     "new": true
    },
    {
     "name": "Kimi K3.1 Pro (standard)",
     "sci": 160.4,
     "n": 111
    }
   ],
   "release_date": "2026-10-23",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": true,
   "new_since": "2026-10-30",
   "badges": [
    "new"
   ],
   "rank": 19
  },
  {
   "key": "gpt 56 sol",
   "name": "GPT-5.6 Sol",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 161.3,
   "n_obs": 295,
   "n_configs": 6,
   "configs": [
    {
     "name": "GPT-5.6 Sol (max)",
     "sci": 161.3,
     "n": 41,
     "headline": true
    },
    {
     "name": "GPT-5.6 Sol (xhigh)",
     "sci": 159.4,
     "n": 18
    },
    {
     "name": "GPT-5.6 Sol (standard)",
     "sci": 159.3,
     "n": 188
    },
    {
     "name": "GPT-5.6 Sol (high)",
     "sci": 158.5,
     "n": 18
    },
    {
     "name": "GPT-5.6 Sol (medium)",
     "sci": 157.5,
     "n": 15
    },
    {
     "name": "GPT-5.6 Sol (low)",
     "sci": 154.4,
     "n": 15
    }
   ],
   "release_date": "2026-07-09",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 20
  },
  {
   "key": "gpt 55 pro pre release",
   "name": "GPT-5.5 Pro Pre Release",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 160.9,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen38 max 0902",
   "name": "Qwen3.8-Max-0902",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 160.8,
   "n_obs": 28,
   "n_configs": 2,
   "configs": [
    {
     "name": "Qwen3.8-Max-0902 (standard)",
     "sci": 160.8,
     "n": 21,
     "headline": true
    },
    {
     "name": "Qwen3.8-Max-0902 (xhigh)",
     "sci": 156.6,
     "n": 7
    }
   ],
   "release_date": "2026-09-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 21
  },
  {
   "key": "muse wonder full",
   "name": "Muse Wonder Full",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 160.8,
   "n_obs": 121,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-10-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 22
  },
  {
   "key": "atria dawn preview",
   "name": "Atria Dawn Preview",
   "organization": "Shanghai AI Laboratory",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 160.8,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "lab-reported only",
    "thin evidence"
   ],
   "rank": 23
  },
  {
   "key": "muse wonder",
   "name": "Muse Wonder",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 160.4,
   "n_obs": 143,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-10-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 24
  },
  {
   "key": "gpt 56 terra",
   "name": "GPT-5.6 Terra",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 160.1,
   "n_obs": 250,
   "n_configs": 7,
   "configs": [
    {
     "name": "GPT-5.6 Terra (max)",
     "sci": 160.1,
     "n": 30,
     "headline": true
    },
    {
     "name": "GPT-5.6 Terra (standard)",
     "sci": 157.5,
     "n": 157
    },
    {
     "name": "GPT-5.6 Terra (xhigh)",
     "sci": 157.3,
     "n": 15
    },
    {
     "name": "GPT-5.6 Terra (high)",
     "sci": 156.3,
     "n": 13
    },
    {
     "name": "GPT-5.6 Terra (medium)",
     "sci": 152.8,
     "n": 13
    },
    {
     "name": "GPT-5.6 Terra (low)",
     "sci": 150.3,
     "n": 14
    },
    {
     "name": "GPT-5.6 Terra (effort unspecified)",
     "sci": 146.0,
     "n": 8
    }
   ],
   "release_date": "2026-07-09",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 25
  },
  {
   "key": "muse wonder 10",
   "name": "Muse Wonder 1.0",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 160.1,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-10-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 26
  },
  {
   "key": "gpt 55 pre release",
   "name": "GPT-5.5 Pre Release",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 159.8,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 54 pro",
   "name": "GPT-5.4 Pro",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 159.7,
   "n_obs": 14,
   "n_configs": 3,
   "configs": [
    {
     "name": "GPT-5.4 Pro (xhigh)",
     "sci": 159.7,
     "n": 9,
     "headline": true
    },
    {
     "name": "GPT-5.4 Pro (standard)",
     "sci": 159.5,
     "n": 5
    }
   ],
   "release_date": "2026-03-05",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 27
  },
  {
   "key": "gpt 55",
   "name": "GPT-5.5",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 159.7,
   "n_obs": 310,
   "n_configs": 5,
   "configs": [
    {
     "name": "GPT-5.5 (xhigh)",
     "sci": 159.7,
     "n": 60,
     "headline": true
    },
    {
     "name": "GPT-5.5 (high)",
     "sci": 157.5,
     "n": 33
    },
    {
     "name": "GPT-5.5 (standard)",
     "sci": 156.6,
     "n": 174
    },
    {
     "name": "GPT-5.5 (medium)",
     "sci": 155.8,
     "n": 30
    },
    {
     "name": "GPT-5.5 (low)",
     "sci": 153.0,
     "n": 13
    }
   ],
   "release_date": "2026-04-23",
   "cost_usd": 360.69161,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 28
  },
  {
   "key": "claude haiku 55",
   "name": "Claude Haiku 5.5",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 159.4,
   "n_obs": 197,
   "n_configs": 6,
   "configs": [
    {
     "name": "Claude Haiku 5.5 (max)",
     "sci": 159.4,
     "n": 44,
     "headline": true
    },
    {
     "name": "Claude Haiku 5.5 (standard)",
     "sci": 158.3,
     "n": 119
    },
    {
     "name": "Claude Haiku 5.5 (xhigh)",
     "sci": 157.6,
     "n": 9
    },
    {
     "name": "Claude Haiku 5.5 (high)",
     "sci": 153.9,
     "n": 8
    },
    {
     "name": "Claude Haiku 5.5 (medium)",
     "sci": 150.1,
     "n": 8
    },
    {
     "name": "Claude Haiku 5.5 (low)",
     "sci": 148.3,
     "n": 9
    }
   ],
   "release_date": "2026-10-07",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 29
  },
  {
   "key": "gemini 38 flash",
   "name": "Gemini 3.8 Flash",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 158.8,
   "n_obs": 274,
   "n_configs": 5,
   "configs": [
    {
     "name": "Gemini 3.8 Flash (unknown)",
     "sci": 160.7,
     "n": 5,
     "thin": true
    },
    {
     "name": "Gemini 3.8 Flash (medium)",
     "sci": 158.8,
     "n": 15,
     "headline": true
    },
    {
     "name": "Gemini 3.8 Flash (high)",
     "sci": 158.0,
     "n": 27
    },
    {
     "name": "Gemini 3.8 Flash (standard)",
     "sci": 157.9,
     "n": 216
    },
    {
     "name": "Gemini 3.8 Flash (low)",
     "sci": 156.9,
     "n": 11
    }
   ],
   "release_date": "2026-09-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 30
  },
  {
   "key": "claude opus 48",
   "name": "Claude Opus 4.8",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 158.7,
   "n_obs": 284,
   "n_configs": 8,
   "configs": [
    {
     "name": "Claude Opus 4.8 (max)",
     "sci": 158.7,
     "n": 45,
     "headline": true
    },
    {
     "name": "Claude Opus 4.8 (effort unspecified)",
     "sci": 157.8,
     "n": 10
    },
    {
     "name": "Claude Opus 4.8 (standard)",
     "sci": 157.7,
     "n": 154
    },
    {
     "name": "Claude Opus 4.8 (xhigh)",
     "sci": 157.6,
     "n": 20
    },
    {
     "name": "Claude Opus 4.8 (high)",
     "sci": 157.6,
     "n": 24
    },
    {
     "name": "Claude Opus 4.8 (Adaptive Reasoning, Max Effort)",
     "sci": 157.5,
     "n": 6
    },
    {
     "name": "Claude Opus 4.8 (medium)",
     "sci": 156.6,
     "n": 19
    },
    {
     "name": "Claude Opus 4.8 (low)",
     "sci": 156.0,
     "n": 6
    }
   ],
   "release_date": "2026-05-28",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 31
  },
  {
   "key": "grok 4 heavy",
   "name": "Grok-4 Heavy",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 158.6,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 54",
   "name": "GPT-5.4",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 158.3,
   "n_obs": 193,
   "n_configs": 7,
   "configs": [
    {
     "name": "GPT-5.4 (xhigh)",
     "sci": 158.3,
     "n": 24,
     "headline": true
    },
    {
     "name": "GPT-5.4 (xhigh)",
     "sci": 157.6,
     "n": 23
    },
    {
     "name": "GPT-5.4 (medium)",
     "sci": 156.0,
     "n": 5
    },
    {
     "name": "GPT-5.4 (high)",
     "sci": 155.5,
     "n": 24
    },
    {
     "name": "GPT-5.4 (standard)",
     "sci": 153.8,
     "n": 112
    },
    {
     "name": "GPT-5.4 (low)",
     "sci": 151.4,
     "n": 5
    }
   ],
   "release_date": "2026-03-05",
   "cost_usd": 271.74131,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 32
  },
  {
   "key": "grok 47",
   "name": "Grok 4.7",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 158.3,
   "n_obs": 244,
   "n_configs": 4,
   "configs": [
    {
     "name": "Grok 4.7 (high)",
     "sci": 158.3,
     "n": 12,
     "headline": true
    },
    {
     "name": "Grok 4.7 (standard)",
     "sci": 157.7,
     "n": 196
    },
    {
     "name": "Grok 4.7 (xhigh)",
     "sci": 156.9,
     "n": 29
    },
    {
     "name": "Grok 4.7 (low)",
     "sci": 151.4,
     "n": 7
    }
   ],
   "release_date": "2026-09-21",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 33
  },
  {
   "key": "nex n25",
   "name": "Nex-N2.5",
   "organization": "Nex-AGI",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 158.2,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-08",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "lab-reported only",
    "thin evidence"
   ],
   "rank": 34
  },
  {
   "key": "glm 53",
   "name": "GLM-5.3",
   "organization": "Zhipu AI",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 158.1,
   "n_obs": 236,
   "n_configs": 2,
   "configs": [
    {
     "name": "GLM-5.3 (standard)",
     "sci": 158.1,
     "n": 229,
     "headline": true
    },
    {
     "name": "GLM-5.3 (low)",
     "sci": 156.8,
     "n": 7
    }
   ],
   "release_date": "2026-08-14",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 35
  },
  {
   "key": "qwen38 max",
   "name": "Qwen3.8 Max",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 158.0,
   "n_obs": 192,
   "n_configs": 2,
   "configs": [
    {
     "name": "Qwen3.8 Max (standard)",
     "sci": 158.0,
     "n": 179,
     "headline": true
    },
    {
     "name": "Qwen3.8 Max (xhigh)",
     "sci": 157.2,
     "n": 13
    }
   ],
   "release_date": "2026-08-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 36
  },
  {
   "key": "gpt 52 pro",
   "name": "GPT-5.2 Pro",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 158.0,
   "n_obs": 10,
   "n_configs": 2,
   "configs": [
    {
     "name": "GPT-5.2 Pro (xhigh)",
     "sci": 158.0,
     "n": 5,
     "headline": true
    },
    {
     "name": "GPT-5.2 Pro (standard)",
     "sci": 154.3,
     "n": 5
    }
   ],
   "release_date": "2025-12-11",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 37
  },
  {
   "key": "muse spark 13",
   "name": "Muse Spark 1.3",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 157.9,
   "n_obs": 177,
   "n_configs": 2,
   "configs": [
    {
     "name": "Muse Spark 1.3 (max)",
     "sci": 157.9,
     "n": 158,
     "headline": true
    },
    {
     "name": "Muse Spark 1.3 (xhigh)",
     "sci": 157.1,
     "n": 19
    }
   ],
   "release_date": "2026-09-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 38
  },
  {
   "key": "gemini 37 flash",
   "name": "Gemini 3.7 Flash",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 157.9,
   "n_obs": 182,
   "n_configs": 4,
   "configs": [
    {
     "name": "Gemini 3.7 Flash (high)",
     "sci": 157.9,
     "n": 23,
     "headline": true
    },
    {
     "name": "Gemini 3.7 Flash (medium)",
     "sci": 157.7,
     "n": 11
    },
    {
     "name": "Gemini 3.7 Flash (standard)",
     "sci": 157.4,
     "n": 138
    },
    {
     "name": "Gemini 3.7 Flash (low)",
     "sci": 155.9,
     "n": 10
    }
   ],
   "release_date": "2026-08-13",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 39
  },
  {
   "key": "grok 46",
   "name": "Grok 4.6",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 157.9,
   "n_obs": 214,
   "n_configs": 5,
   "configs": [
    {
     "name": "Grok 4.6 (high)",
     "sci": 157.9,
     "n": 22,
     "headline": true
    },
    {
     "name": "Grok 4.6 (standard)",
     "sci": 157.7,
     "n": 145
    },
    {
     "name": "Grok 4.6 (medium)",
     "sci": 157.4,
     "n": 11
    },
    {
     "name": "Grok 4.6 (xhigh)",
     "sci": 157.3,
     "n": 25
    },
    {
     "name": "Grok 4.6 (low)",
     "sci": 153.0,
     "n": 11
    }
   ],
   "release_date": "2026-08-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 40
  },
  {
   "key": "kimi k3",
   "name": "Kimi K3",
   "organization": "Moonshot AI",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 157.8,
   "n_obs": 303,
   "n_configs": 5,
   "configs": [
    {
     "name": "Kimi K3 (effort unspecified)",
     "sci": 158.5,
     "n": 6,
     "thin": true
    },
    {
     "name": "Kimi K3 (max)",
     "sci": 157.8,
     "n": 47,
     "headline": true
    },
    {
     "name": "Kimi K3 (standard)",
     "sci": 157.8,
     "n": 229
    },
    {
     "name": "Kimi K3 (high)",
     "sci": 156.6,
     "n": 8
    },
    {
     "name": "Kimi K3 (low)",
     "sci": 151.5,
     "n": 13
    }
   ],
   "release_date": "2026-07-16",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 41
  },
  {
   "key": "deepseek v41 flash 1017",
   "name": "DeepSeek V4.1 Flash 1017",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 157.8,
   "n_obs": 116,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-10-17",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 42
  },
  {
   "key": "deepseek v41 flash",
   "name": "DeepSeek V4.1 Flash",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 157.8,
   "n_obs": 223,
   "n_configs": 3,
   "configs": [
    {
     "name": "DeepSeek V4.1 Flash (Reasoning, Max Effort)",
     "sci": 157.8,
     "n": 35,
     "headline": true
    },
    {
     "name": "DeepSeek V4.1 Flash (high)",
     "sci": 156.6,
     "n": 8
    },
    {
     "name": "DeepSeek V4.1 Flash (standard)",
     "sci": 156.4,
     "n": 180
    }
   ],
   "release_date": "2026-09-10",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 43
  },
  {
   "key": "iquest q1",
   "name": "IQuest-Q1",
   "organization": "IQuest",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 157.8,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-29",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "lab-reported only",
    "thin evidence"
   ],
   "rank": 44
  },
  {
   "key": "gpt 56 luna",
   "name": "GPT-5.6 Luna",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 157.6,
   "n_obs": 258,
   "n_configs": 7,
   "configs": [
    {
     "name": "GPT-5.6 Luna (max)",
     "sci": 157.6,
     "n": 27,
     "headline": true
    },
    {
     "name": "GPT-5.6 Luna (standard)",
     "sci": 155.9,
     "n": 165
    },
    {
     "name": "GPT-5.6 Luna (xhigh)",
     "sci": 155.0,
     "n": 15
    },
    {
     "name": "GPT-5.6 Luna (high)",
     "sci": 153.0,
     "n": 12
    },
    {
     "name": "GPT-5.6 Luna (medium)",
     "sci": 149.2,
     "n": 15
    },
    {
     "name": "GPT-5.6 Luna (low)",
     "sci": 146.0,
     "n": 14
    },
    {
     "name": "GPT-5.6 Luna (effort unspecified)",
     "sci": 144.1,
     "n": 10
    }
   ],
   "release_date": "2026-07-09",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 45
  },
  {
   "key": "intern s2397b",
   "name": "Intern-S2-397B",
   "organization": "Shanghai AI Laboratory",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 157.6,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-13",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mimo v26 pro",
   "name": "MiMo-V2.6-Pro",
   "organization": "Xiaomi",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 157.5,
   "n_obs": 188,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 46
  },
  {
   "key": "gpt 6 luna",
   "name": "GPT-6 Luna",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 157.5,
   "n_obs": 259,
   "n_configs": 6,
   "configs": [
    {
     "name": "GPT-6 Luna (max)",
     "sci": 157.5,
     "n": 24,
     "headline": true
    },
    {
     "name": "GPT-6 Luna (standard)",
     "sci": 155.4,
     "n": 200
    },
    {
     "name": "GPT-6 Luna (xhigh)",
     "sci": 154.4,
     "n": 10
    },
    {
     "name": "GPT-6 Luna (high)",
     "sci": 153.0,
     "n": 9
    },
    {
     "name": "GPT-6 Luna (medium)",
     "sci": 150.8,
     "n": 8
    },
    {
     "name": "GPT-6 Luna (low)",
     "sci": 146.0,
     "n": 8
    }
   ],
   "release_date": "2026-09-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 47
  },
  {
   "key": "gpt 56 sol pro",
   "name": "GPT-5.6 Sol Pro",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 157.4,
   "n_obs": 16,
   "n_configs": 2,
   "configs": [
    {
     "name": "GPT-5.6 Sol Pro (max)",
     "sci": 162.2,
     "n": 5,
     "thin": true
    },
    {
     "name": "GPT-5.6 Sol Pro (standard)",
     "sci": 157.4,
     "n": 11,
     "headline": true
    }
   ],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 48
  },
  {
   "key": "inkling 2 small",
   "name": "Inkling 2 Small",
   "organization": "Thinking Machines",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 157.2,
   "n_obs": 110,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-11-05",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 49
  },
  {
   "key": "qwen3824t a95b",
   "name": "Qwen3.8 2.4T A95B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 157.1,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 50
  },
  {
   "key": "claude opus 47",
   "name": "Claude Opus 4.7",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 157.1,
   "n_obs": 234,
   "n_configs": 7,
   "configs": [
    {
     "name": "Claude Opus 4.7 (max)",
     "sci": 157.1,
     "n": 20,
     "headline": true
    },
    {
     "name": "Claude Opus 4.7 (standard)",
     "sci": 156.5,
     "n": 149
    },
    {
     "name": "Claude Opus 4.7 (high)",
     "sci": 156.4,
     "n": 21
    },
    {
     "name": "Claude Opus 4.7 (Adaptive Reasoning, Max Effort)",
     "sci": 156.0,
     "n": 5
    },
    {
     "name": "Claude Opus 4.7 (xhigh)",
     "sci": 155.9,
     "n": 23
    },
    {
     "name": "Claude Opus 4.7 (medium)",
     "sci": 155.3,
     "n": 16
    }
   ],
   "release_date": "2026-04-16",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 51
  },
  {
   "key": "gemini 31 pro preview",
   "name": "Gemini 3.1 Pro Preview",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 156.9,
   "n_obs": 22,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 52
  },
  {
   "key": "step 5 preview",
   "name": "Step 5 Preview",
   "organization": "StepFun",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 156.8,
   "n_obs": 107,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 53
  },
  {
   "key": "glm 53 flash",
   "name": "GLM-5.3-Flash",
   "organization": "Zhipu AI",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 156.7,
   "n_obs": 195,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-25",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 54
  },
  {
   "key": "mimo v26 flash",
   "name": "MiMo-V2.6-Flash",
   "organization": "Xiaomi",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 156.7,
   "n_obs": 178,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 55
  },
  {
   "key": "claude sonnet 5",
   "name": "Claude Sonnet 5",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 156.6,
   "n_obs": 235,
   "n_configs": 9,
   "configs": [
    {
     "name": "Claude Sonnet 5 (effort unspecified)",
     "sci": 157.7,
     "n": 6,
     "thin": true
    },
    {
     "name": "Claude Sonnet 5 (Adaptive Reasoning, Max Effort)",
     "sci": 156.7,
     "n": 5,
     "thin": true
    },
    {
     "name": "Claude Sonnet 5 (max)",
     "sci": 156.6,
     "n": 27,
     "headline": true
    },
    {
     "name": "Claude Sonnet 5 (standard)",
     "sci": 156.4,
     "n": 150
    },
    {
     "name": "Claude Sonnet 5 (xhigh)",
     "sci": 154.9,
     "n": 23
    },
    {
     "name": "Claude Sonnet 5 (high)",
     "sci": 153.9,
     "n": 9
    },
    {
     "name": "Claude Sonnet 5 (medium)",
     "sci": 152.3,
     "n": 8
    },
    {
     "name": "Claude Sonnet 5 (low)",
     "sci": 149.2,
     "n": 7
    }
   ],
   "release_date": "2026-06-30",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 56
  },
  {
   "key": "deepseek v4 pro 0813",
   "name": "DeepSeek V4 Pro 0813",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 156.6,
   "n_obs": 208,
   "n_configs": 3,
   "configs": [
    {
     "name": "DeepSeek V4 Pro 0813 (Reasoning, Max Effort)",
     "sci": 156.6,
     "n": 19,
     "headline": true
    },
    {
     "name": "DeepSeek V4 Pro 0813 (high)",
     "sci": 156.6,
     "n": 5
    },
    {
     "name": "DeepSeek V4 Pro 0813 (standard)",
     "sci": 155.9,
     "n": 184
    }
   ],
   "release_date": "2026-08-13",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 57
  },
  {
   "key": "muse spark 11",
   "name": "Muse Spark 1.1",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 156.4,
   "n_obs": 130,
   "n_configs": 3,
   "configs": [
    {
     "name": "Muse Spark 1.1 (standard)",
     "sci": 156.4,
     "n": 109,
     "headline": true
    },
    {
     "name": "Muse Spark 1.1 (xhigh)",
     "sci": 155.3,
     "n": 9
    },
    {
     "name": "Muse Spark 1.1 (high)",
     "sci": 155.1,
     "n": 12
    }
   ],
   "release_date": "2026-07-09",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 58
  },
  {
   "key": "muse spark open",
   "name": "Muse Spark Open",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 156.3,
   "n_obs": 95,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-10-15",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 59
  },
  {
   "key": "minimax m31 flash",
   "name": "MiniMax-M3.1-Flash",
   "organization": "MiniMax",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 156.3,
   "n_obs": 91,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-10-19",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 60
  },
  {
   "key": "seed 21 turbo",
   "name": "Seed 2.1 Turbo",
   "organization": "ByteDance",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 156.3,
   "n_obs": 35,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-06-24",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 61
  },
  {
   "key": "hy4 preview",
   "name": "Hy4 preview",
   "organization": "Tencent",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 156.3,
   "n_obs": 202,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-28",
   "cost_usd": null,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 62
  },
  {
   "key": "nex n25 pro",
   "name": "Nex-N2.5-Pro",
   "organization": "Nex-AGI",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 156.3,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-08",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "lab-reported only",
    "thin evidence"
   ],
   "rank": 63
  },
  {
   "key": "gpt 52",
   "name": "GPT-5.2",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 156.2,
   "n_obs": 145,
   "n_configs": 7,
   "configs": [
    {
     "name": "GPT-5.2 (xhigh)",
     "sci": 156.2,
     "n": 18,
     "headline": true
    },
    {
     "name": "GPT-5.2 (high)",
     "sci": 153.6,
     "n": 28
    },
    {
     "name": "GPT-5.2 (medium)",
     "sci": 153.0,
     "n": 26
    },
    {
     "name": "GPT-5.2 (standard)",
     "sci": 150.8,
     "n": 50
    },
    {
     "name": "GPT-5.2 (low)",
     "sci": 148.1,
     "n": 23
    }
   ],
   "release_date": "2025-12-11",
   "cost_usd": 225.43901275000002,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 64
  },
  {
   "key": "deepseek v4 flash vision exp",
   "name": "DeepSeek V4 Flash Vision Exp",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 156.2,
   "n_obs": 15,
   "n_configs": 2,
   "configs": [
    {
     "name": "DeepSeek V4 Flash Vision Exp (Reasoning, Max Effort)",
     "sci": 156.2,
     "n": 9,
     "headline": true
    },
    {
     "name": "DeepSeek V4 Flash Vision Exp (standard)",
     "sci": 155.4,
     "n": 6
    }
   ],
   "release_date": "2026-08-21",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 65
  },
  {
   "key": "muse spark 12",
   "name": "Muse Spark 1.2",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 156.2,
   "n_obs": 142,
   "n_configs": 2,
   "configs": [
    {
     "name": "Muse Spark 1.2 (standard)",
     "sci": 156.2,
     "n": 127,
     "headline": true
    },
    {
     "name": "Muse Spark 1.2 (xhigh)",
     "sci": 155.2,
     "n": 15
    }
   ],
   "release_date": "2026-08-05",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 66
  },
  {
   "key": "grok 45",
   "name": "Grok 4.5",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 156.0,
   "n_obs": 170,
   "n_configs": 3,
   "configs": [
    {
     "name": "Grok 4.5 (effort unspecified)",
     "sci": 159.7,
     "n": 7,
     "thin": true
    },
    {
     "name": "Grok 4.5 (high)",
     "sci": 156.0,
     "n": 24,
     "headline": true
    },
    {
     "name": "Grok 4.5 (standard)",
     "sci": 156.0,
     "n": 139
    }
   ],
   "release_date": "2026-07-16",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 67
  },
  {
   "key": "qwen38 flash",
   "name": "Qwen3.8-Flash",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 155.8,
   "n_obs": 12,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-26",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "lab-reported only"
   ],
   "rank": 68
  },
  {
   "key": "mistral large 4",
   "name": "Mistral Large 4",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "hybrid",
   "sci": 155.8,
   "n_obs": 114,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-10-06",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 69
  },
  {
   "key": "deepseek v4 flash 0731",
   "name": "DeepSeek V4 Flash 0731",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 155.7,
   "n_obs": 167,
   "n_configs": 4,
   "configs": [
    {
     "name": "DeepSeek V4 Flash 0731 (max)",
     "sci": 155.7,
     "n": 15,
     "headline": true
    },
    {
     "name": "DeepSeek V4 Flash 0731 (high)",
     "sci": 155.3,
     "n": 5
    },
    {
     "name": "DeepSeek V4 Flash 0731 (standard)",
     "sci": 154.1,
     "n": 147
    }
   ],
   "release_date": "2026-07-31",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 70
  },
  {
   "key": "gemini 36 flash",
   "name": "Gemini 3.6 Flash",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 155.6,
   "n_obs": 155,
   "n_configs": 4,
   "configs": [
    {
     "name": "Gemini 3.6 Flash (high)",
     "sci": 155.6,
     "n": 25,
     "headline": true
    },
    {
     "name": "Gemini 3.6 Flash (standard)",
     "sci": 155.5,
     "n": 125
    },
    {
     "name": "Gemini 3.6 Flash (low)",
     "sci": 153.3,
     "n": 5
    }
   ],
   "release_date": "2026-07-21",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 71
  },
  {
   "key": "seed 21 pro",
   "name": "Seed 2.1 Pro",
   "organization": "ByteDance",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 155.6,
   "n_obs": 32,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-06-24",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 72
  },
  {
   "key": "gemini 35 flash",
   "name": "Gemini 3.5 Flash",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 155.5,
   "n_obs": 203,
   "n_configs": 5,
   "configs": [
    {
     "name": "Gemini 3.5 Flash (high)",
     "sci": 155.5,
     "n": 39,
     "headline": true
    },
    {
     "name": "Gemini 3.5 Flash (standard)",
     "sci": 154.9,
     "n": 151
    },
    {
     "name": "Gemini 3.5 Flash (medium)",
     "sci": 153.2,
     "n": 7
    },
    {
     "name": "Gemini 3.5 Flash (minimal)",
     "sci": 147.5,
     "n": 6
    }
   ],
   "release_date": "2026-05-19",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 73
  },
  {
   "key": "claude opus 46",
   "name": "Claude Opus 4.6",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 155.4,
   "n_obs": 89,
   "n_configs": 7,
   "configs": [
    {
     "name": "Claude Opus 4.6 (max)",
     "sci": 155.4,
     "n": 30,
     "headline": true
    },
    {
     "name": "Claude Opus 4.6 (standard)",
     "sci": 154.8,
     "n": 50
    },
    {
     "name": "Claude Opus 4.6 (high)",
     "sci": 154.0,
     "n": 9
    }
   ],
   "release_date": "2026-02-05",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 74
  },
  {
   "key": "gpt 53 codex",
   "name": "GPT-5.3 Codex",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 155.2,
   "n_obs": 62,
   "n_configs": 3,
   "configs": [
    {
     "name": "GPT-5.3 Codex (standard)",
     "sci": 155.2,
     "n": 21,
     "headline": true
    },
    {
     "name": "GPT-5.3 Codex (high)",
     "sci": 152.1,
     "n": 15
    },
    {
     "name": "GPT-5.3 Codex (xhigh)",
     "sci": 151.8,
     "n": 26
    }
   ],
   "release_date": "2026-02-05",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 75
  },
  {
   "key": "agnes 25 pro beta",
   "name": "Agnes 2.5 Pro Beta",
   "organization": "Agnes AI",
   "country": "Singapore",
   "reasoning_status": "hybrid",
   "sci": 154.9,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 76
  },
  {
   "key": "glm 52",
   "name": "GLM-5.2",
   "organization": "Zhipu AI",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 154.9,
   "n_obs": 171,
   "n_configs": 5,
   "configs": [
    {
     "name": "GLM-5.2 (standard)",
     "sci": 154.9,
     "n": 155,
     "headline": true
    },
    {
     "name": "GLM-5.2 (effort unspecified)",
     "sci": 153.6,
     "n": 6
    },
    {
     "name": "GLM-5.2 (high)",
     "sci": 153.2,
     "n": 5
    },
    {
     "name": "GLM-5.2 (Non-reasoning Effort)",
     "sci": 138.9,
     "n": 5
    }
   ],
   "release_date": "2026-06-16",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 77
  },
  {
   "key": "claude sonnet 46",
   "name": "Claude Sonnet 4.6",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 154.7,
   "n_obs": 166,
   "n_configs": 7,
   "configs": [
    {
     "name": "Claude Sonnet 4.6 (standard)",
     "sci": 154.7,
     "n": 136,
     "headline": true
    },
    {
     "name": "Claude Sonnet 4.6 (high)",
     "sci": 154.6,
     "n": 8
    },
    {
     "name": "Claude Sonnet 4.6 (max)",
     "sci": 154.4,
     "n": 16
    },
    {
     "name": "Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)",
     "sci": 153.5,
     "n": 6
    }
   ],
   "release_date": "2026-02-17",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 78
  },
  {
   "key": "motif 3",
   "name": "Motif 3",
   "organization": "Unknown",
   "country": "KR",
   "reasoning_status": "unknown",
   "sci": 154.3,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3827b",
   "name": "Qwen3.8 27B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 154.3,
   "n_obs": 153,
   "n_configs": 4,
   "configs": [
    {
     "name": "Qwen3.8 27B (standard)",
     "sci": 154.3,
     "n": 128,
     "headline": true
    },
    {
     "name": "Qwen3.8 27B (xhigh)",
     "sci": 154.3,
     "n": 9
    },
    {
     "name": "Qwen3.8 27B (low)",
     "sci": 151.2,
     "n": 8
    },
    {
     "name": "Qwen3.8 27B (medium)",
     "sci": 150.9,
     "n": 8
    }
   ],
   "release_date": "2026-08-14",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 79
  },
  {
   "key": "mercury 3",
   "name": "Mercury 3",
   "organization": "Inception",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 153.8,
   "n_obs": 72,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-10-19",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 80
  },
  {
   "key": "composer 25",
   "name": "Composer 2.5",
   "organization": "Cursor (Anysphere)",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 153.8,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 81
  },
  {
   "key": "motif 3 (beta)",
   "name": "Motif 3 (Beta)",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 153.7,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen37 max",
   "name": "Qwen3.7 Max",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 153.6,
   "n_obs": 143,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-05-19",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 82
  },
  {
   "key": "gemini 3 flash preview",
   "name": "Gemini 3 Flash Preview",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 153.4,
   "n_obs": 34,
   "n_configs": 2,
   "configs": [
    {
     "name": "Gemini 3 Flash Preview (high)",
     "sci": 153.4,
     "n": 19,
     "headline": true
    },
    {
     "name": "Gemini 3 Flash Preview (minimal)",
     "sci": 142.6,
     "n": 15
    }
   ],
   "release_date": "2025-12-17",
   "cost_usd": 65.979814,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 83
  },
  {
   "key": "qwen38 flash next",
   "name": "Qwen3.8-Flash-Next",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 153.4,
   "n_obs": 21,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-26",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 84
  },
  {
   "key": "k2 horizon 375b a23b",
   "name": "K2 Horizon 375B-A23B",
   "organization": "MBZUAI",
   "country": "UAE",
   "reasoning_status": "hybrid",
   "sci": 153.3,
   "n_obs": 14,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-03",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 85
  },
  {
   "key": "deepseek v4 flash 0423",
   "name": "DeepSeek V4 Flash 0423",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 153.0,
   "n_obs": 30,
   "n_configs": 4,
   "configs": [
    {
     "name": "DeepSeek V4 Flash 0423 (max)",
     "sci": 153.0,
     "n": 14,
     "headline": true
    },
    {
     "name": "DeepSeek V4 Flash 0423 (Reasoning, Max Effort)",
     "sci": 152.5,
     "n": 5
    },
    {
     "name": "DeepSeek V4 Flash 0423 (Reasoning, High Effort)",
     "sci": 151.3,
     "n": 5
    },
    {
     "name": "DeepSeek V4 Flash 0423 (standard)",
     "sci": 148.5,
     "n": 6
    }
   ],
   "release_date": "2026-04-23",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 86
  },
  {
   "key": "gemini 31 pro",
   "name": "Gemini 3.1 Pro",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 153.0,
   "n_obs": 198,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-02-19",
   "cost_usd": 892.283804,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 87
  },
  {
   "key": "claude opus 45",
   "name": "Claude Opus 4.5",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 152.9,
   "n_obs": 88,
   "n_configs": 6,
   "configs": [
    {
     "name": "Claude Opus 4.5 (standard)",
     "sci": 152.9,
     "n": 14,
     "headline": true
    },
    {
     "name": "Claude Opus 4.5 (standard)",
     "sci": 150.8,
     "n": 5
    },
    {
     "name": "Claude Opus 4.5 (20251101 snapshot)",
     "sci": 149.0,
     "n": 28
    },
    {
     "name": "Claude Opus 4.5 (standard)",
     "sci": 148.0,
     "n": 8
    },
    {
     "name": "Claude Opus 4.5 (high)",
     "sci": 144.8,
     "n": 19
    },
    {
     "name": "Claude Opus 4.5 (low)",
     "sci": 142.2,
     "n": 14
    }
   ],
   "release_date": "2025-11-24",
   "cost_usd": 1391.7497250000001,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 88
  },
  {
   "key": "apodex 11",
   "name": "Apodex 1.1",
   "organization": "Apodex",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 152.8,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-24",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 89
  },
  {
   "key": "gpt 52 codex",
   "name": "GPT-5.2 Codex",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 152.7,
   "n_obs": 23,
   "n_configs": 2,
   "configs": [
    {
     "name": "GPT-5.2 Codex (standard)",
     "sci": 152.7,
     "n": 23,
     "headline": true
    }
   ],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 90
  },
  {
   "key": "minimax m3",
   "name": "MiniMax-M3",
   "organization": "MiniMax",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 152.7,
   "n_obs": 192,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-06-01",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 91
  },
  {
   "key": "deepseek v4 pro",
   "name": "DeepSeek V4 Pro",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 152.6,
   "n_obs": 196,
   "n_configs": 5,
   "configs": [
    {
     "name": "DeepSeek V4 Pro (Reasoning, Max Effort)",
     "sci": 154.1,
     "n": 6,
     "thin": true
    },
    {
     "name": "DeepSeek V4 Pro (standard)",
     "sci": 152.6,
     "n": 158,
     "headline": true
    },
    {
     "name": "DeepSeek V4 Pro (Reasoning, High Effort)",
     "sci": 152.6,
     "n": 5
    },
    {
     "name": "DeepSeek V4 Pro (max)",
     "sci": 152.5,
     "n": 21
    },
    {
     "name": "DeepSeek V4 Pro (high)",
     "sci": 148.9,
     "n": 6
    }
   ],
   "release_date": "2026-04-23",
   "cost_usd": 615.8715906,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 92
  },
  {
   "key": "jt 41 flash 236b a21b",
   "name": "JT-4.1 Flash 236B A21B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 152.6,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "seed 20 pro",
   "name": "Seed 2.0 Pro",
   "organization": "ByteDance",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 152.6,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-02-14",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 3 pro preview",
   "name": "Gemini 3 Pro Preview",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 152.5,
   "n_obs": 41,
   "n_configs": 3,
   "configs": [
    {
     "name": "Gemini 3 Pro Preview (standard)",
     "sci": 152.5,
     "n": 41,
     "headline": true
    }
   ],
   "release_date": "2025-11-18",
   "cost_usd": null,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 93
  },
  {
   "key": "kimi k27 code",
   "name": "Kimi K2.7 Code",
   "organization": "Moonshot AI",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 152.5,
   "n_obs": 59,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-06-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 94
  },
  {
   "key": "nex n2 pro",
   "name": "Nex-N2-Pro",
   "organization": "Nex AGI",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 152.4,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "ernie 50",
   "name": "ERNIE 5.0",
   "organization": "Baidu",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 152.2,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "nex n25 mini",
   "name": "Nex-N2.5-mini",
   "organization": "Nex-AGI",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 152.1,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-08",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "lab-reported only",
    "thin evidence"
   ],
   "rank": 95
  },
  {
   "key": "gpt 54 nano",
   "name": "GPT-5.4 nano",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 152.1,
   "n_obs": 211,
   "n_configs": 7,
   "configs": [
    {
     "name": "GPT-5.4 nano (xhigh)",
     "sci": 152.1,
     "n": 23,
     "headline": true
    },
    {
     "name": "GPT-5.4 nano (high)",
     "sci": 148.6,
     "n": 23
    },
    {
     "name": "GPT-5.4 nano (xhigh)",
     "sci": 148.2,
     "n": 6
    },
    {
     "name": "GPT-5.4 nano (medium)",
     "sci": 146.4,
     "n": 16
    },
    {
     "name": "GPT-5.4 nano (2026-03-17 snapshot)",
     "sci": 145.8,
     "n": 103
    },
    {
     "name": "GPT-5.4 nano (low)",
     "sci": 140.5,
     "n": 18
    },
    {
     "name": "GPT-5.4 nano (Non-Reasoning)",
     "sci": 138.2,
     "n": 22
    }
   ],
   "release_date": "2026-03-17",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 96
  },
  {
   "key": "grok 4200309 v2",
   "name": "Grok 4.20 0309 v2",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 152.0,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 51",
   "name": "GPT-5.1",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 151.9,
   "n_obs": 101,
   "n_configs": 6,
   "configs": [
    {
     "name": "GPT-5.1 (high)",
     "sci": 151.9,
     "n": 27,
     "headline": true
    },
    {
     "name": "GPT-5.1 (medium)",
     "sci": 149.5,
     "n": 23
    },
    {
     "name": "GPT-5.1 (high)",
     "sci": 149.2,
     "n": 7
    },
    {
     "name": "GPT-5.1 (2025-11-13 snapshot)",
     "sci": 146.7,
     "n": 25
    },
    {
     "name": "GPT-5.1 (low)",
     "sci": 145.1,
     "n": 19
    }
   ],
   "release_date": "2025-11-12",
   "cost_usd": 94.73182625,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 97
  },
  {
   "key": "gpt 52 chat",
   "name": "GPT-5.2 Chat",
   "organization": "OpenAI",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 151.9,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "ling 30 flash vl",
   "name": "Ling 3.0 Flash VL",
   "organization": "Ant Group",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 151.9,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 98
  },
  {
   "key": "qwen36 max preview",
   "name": "Qwen3.6 Max Preview",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 151.8,
   "n_obs": 19,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 99
  },
  {
   "key": "agnes 25 pro alpha",
   "name": "Agnes 2.5 Pro Alpha",
   "organization": "Agnes AI",
   "country": "Singapore",
   "reasoning_status": "unknown",
   "sci": 151.7,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 51 thinking",
   "name": "GPT-5.1 Thinking",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 151.7,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-11-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "hy3 (july 2026)",
   "name": "Hy3 (July 2026)",
   "organization": "Tencent",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 151.6,
   "n_obs": 36,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-07-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 100
  },
  {
   "key": "muse spark",
   "name": "Muse Spark",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "unknown",
   "sci": 151.5,
   "n_obs": 23,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 101
  },
  {
   "key": "muse glimmer 30b",
   "name": "Muse Glimmer-30B",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "unknown",
   "sci": 151.4,
   "n_obs": 17,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-10",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 102
  },
  {
   "key": "kimi k26",
   "name": "Kimi K2.6",
   "organization": "Moonshot AI",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 151.3,
   "n_obs": 180,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-04-20",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 103
  },
  {
   "key": "glm 51",
   "name": "GLM-5.1",
   "organization": "Zhipu AI",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 151.3,
   "n_obs": 138,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-04-07",
   "cost_usd": 617.5646028,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 104
  },
  {
   "key": "gpt 5 pro",
   "name": "GPT-5 Pro",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 151.2,
   "n_obs": 25,
   "n_configs": 2,
   "configs": [
    {
     "name": "GPT-5 Pro (high)",
     "sci": 154.4,
     "n": 7,
     "thin": true
    },
    {
     "name": "GPT-5 Pro (2025-10-06 snapshot)",
     "sci": 151.2,
     "n": 18,
     "headline": true
    }
   ],
   "release_date": "2025-10-06",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 105
  },
  {
   "key": "quasar 438b",
   "name": "Quasar 438B",
   "organization": "Multiverse Computing",
   "country": "Spain",
   "reasoning_status": "reasoning",
   "sci": 151.2,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 106
  },
  {
   "key": "kat coder pro v2",
   "name": "KAT Coder Pro V2",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 151.2,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 107
  },
  {
   "key": "grok build 010616",
   "name": "Grok Build 0.1 0616",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 151.1,
   "n_obs": 18,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 108
  },
  {
   "key": "gpt 54 mini",
   "name": "GPT-5.4 mini",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 151.0,
   "n_obs": 224,
   "n_configs": 6,
   "configs": [
    {
     "name": "GPT-5.4 mini (xhigh)",
     "sci": 151.0,
     "n": 19,
     "headline": true
    },
    {
     "name": "GPT-5.4 mini (xhigh)",
     "sci": 150.8,
     "n": 14
    },
    {
     "name": "GPT-5.4 mini (high)",
     "sci": 149.0,
     "n": 20
    },
    {
     "name": "GPT-5.4 mini (2026-03-17 snapshot)",
     "sci": 145.7,
     "n": 137
    },
    {
     "name": "GPT-5.4 mini (medium)",
     "sci": 145.6,
     "n": 16
    },
    {
     "name": "GPT-5.4 mini (low)",
     "sci": 139.6,
     "n": 18
    }
   ],
   "release_date": "2026-03-17",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 109
  },
  {
   "key": "mimo v25 pro",
   "name": "MiMo-V2.5-Pro",
   "organization": "Xiaomi",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 151.0,
   "n_obs": 114,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-04-23",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 110
  },
  {
   "key": "gemini 3 pro",
   "name": "Gemini 3 Pro",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 150.8,
   "n_obs": 14,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-11-18",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 111
  },
  {
   "key": "gpt 53 chat",
   "name": "GPT-5.3 Chat",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 150.7,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-03-04",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 112
  },
  {
   "key": "glm 53 flashx",
   "name": "GLM-5.3-FlashX",
   "organization": "Zhipu AI",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 150.7,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 51 codex",
   "name": "GPT-5.1 Codex",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 150.5,
   "n_obs": 47,
   "n_configs": 3,
   "configs": [
    {
     "name": "GPT-5.1 Codex (max)",
     "sci": 150.5,
     "n": 21,
     "headline": true
    },
    {
     "name": "GPT-5.1 Codex (high)",
     "sci": 149.9,
     "n": 5
    },
    {
     "name": "GPT-5.1 Codex (standard)",
     "sci": 149.7,
     "n": 21
    }
   ],
   "release_date": "2025-11-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 113
  },
  {
   "key": "qwen37 plus",
   "name": "Qwen3.7 Plus",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 150.5,
   "n_obs": 130,
   "n_configs": 2,
   "configs": [
    {
     "name": "Qwen3.7 Plus (standard)",
     "sci": 150.5,
     "n": 130,
     "headline": true
    }
   ],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 114
  },
  {
   "key": "grok 43",
   "name": "Grok 4.3",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 150.4,
   "n_obs": 162,
   "n_configs": 5,
   "configs": [
    {
     "name": "Grok 4.3 (high)",
     "sci": 150.4,
     "n": 17,
     "headline": true
    },
    {
     "name": "Grok 4.3 (standard)",
     "sci": 147.2,
     "n": 140
    },
    {
     "name": "Grok 4.3 (unknown)",
     "sci": 143.4,
     "n": 5
    }
   ],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 115
  },
  {
   "key": "qwen36 plus",
   "name": "Qwen3.6-Plus",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 150.3,
   "n_obs": 151,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-03-31",
   "cost_usd": 482.650482,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 116
  },
  {
   "key": "mimo v2 omni 0327",
   "name": "MiMo-V2-Omni-0327",
   "organization": "Xiaomi",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 150.2,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 35 flash lite",
   "name": "Gemini 3.5 Flash-Lite",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 150.1,
   "n_obs": 121,
   "n_configs": 4,
   "configs": [
    {
     "name": "Gemini 3.5 Flash-Lite (standard)",
     "sci": 150.1,
     "n": 109,
     "headline": true
    },
    {
     "name": "Gemini 3.5 Flash-Lite (high)",
     "sci": 146.6,
     "n": 12
    }
   ],
   "release_date": "2026-07-21",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 117
  },
  {
   "key": "grok build 01",
   "name": "Grok Build 0.1",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "unknown",
   "sci": 150.1,
   "n_obs": 20,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 118
  },
  {
   "key": "gpt 5 codex",
   "name": "GPT-5 Codex",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 150.1,
   "n_obs": 12,
   "n_configs": 2,
   "configs": [
    {
     "name": "GPT-5 Codex (high)",
     "sci": 150.1,
     "n": 5,
     "headline": true
    },
    {
     "name": "GPT-5 Codex (standard)",
     "sci": 147.9,
     "n": 7
    }
   ],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 119
  },
  {
   "key": "qwen35397b a17b",
   "name": "Qwen3.5-397B-A17B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 150.0,
   "n_obs": 25,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-02-16",
   "cost_usd": 186.3738378,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 120
  },
  {
   "key": "gpt 5",
   "name": "GPT-5",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 150.0,
   "n_obs": 85,
   "n_configs": 7,
   "configs": [
    {
     "name": "GPT-5 (high)",
     "sci": 150.0,
     "n": 34,
     "headline": true
    },
    {
     "name": "GPT-5 (medium)",
     "sci": 149.3,
     "n": 16
    },
    {
     "name": "GPT-5 (2025-08-07 snapshot)",
     "sci": 148.9,
     "n": 20
    },
    {
     "name": "GPT-5 (low)",
     "sci": 147.4,
     "n": 6
    },
    {
     "name": "GPT-5 (minimal)",
     "sci": 138.2,
     "n": 9
    }
   ],
   "release_date": "2025-08-07",
   "cost_usd": 551.724095,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 121
  },
  {
   "key": "mimo v25",
   "name": "MiMo-V2.5",
   "organization": "Xiaomi",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 150.0,
   "n_obs": 107,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-04-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 122
  },
  {
   "key": "step 37 flash",
   "name": "Step 3.7 Flash",
   "organization": "StepFun",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 150.0,
   "n_obs": 16,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-06-10",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 123
  },
  {
   "key": "inkling",
   "name": "Inkling",
   "organization": "Thinking Machines",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 149.9,
   "n_obs": 189,
   "n_configs": 2,
   "configs": [
    {
     "name": "Inkling (xhigh)",
     "sci": 149.9,
     "n": 14,
     "headline": true
    },
    {
     "name": "Inkling (standard)",
     "sci": 148.8,
     "n": 175
    }
   ],
   "release_date": "2026-07-15",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 124
  },
  {
   "key": "mimo v2 flash (feb 2026)",
   "name": "MiMo-V2-Flash (Feb 2026)",
   "organization": "Xiaomi",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 149.9,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "seed 20 lite",
   "name": "Seed 2.0 Lite",
   "organization": "ByteDance",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 149.9,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-02-14",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "muse glimmer",
   "name": "Muse Glimmer",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "unknown",
   "sci": 149.9,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-10",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 125
  },
  {
   "key": "seed 20 code",
   "name": "Seed-2.0-Code",
   "organization": "ByteDance",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 149.9,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-02-14",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 51 instant",
   "name": "GPT-5.1 Instant",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 149.8,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-11-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "ling 31 flash",
   "name": "Ling 3.1 Flash",
   "organization": "Ant Group",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 149.8,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-30",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 126
  },
  {
   "key": "k2 horizon mova 36b a4b",
   "name": "K2 Horizon MoVA 36B-A4B",
   "organization": "MBZUAI",
   "country": "UAE",
   "reasoning_status": "hybrid",
   "sci": 149.6,
   "n_obs": 8,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-03",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 127
  },
  {
   "key": "qwen35 plus 20260215",
   "name": "Qwen3.5 Plus 2026-02-15",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 149.6,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "glm 5",
   "name": "GLM-5",
   "organization": "Zhipu AI",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 149.6,
   "n_obs": 38,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-02-11",
   "cost_usd": 239.9912732,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 128
  },
  {
   "key": "nova 2 omni",
   "name": "Nova 2 Omni",
   "organization": "Amazon",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 149.6,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3527b",
   "name": "Qwen3.5 27B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 149.4,
   "n_obs": 26,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 129
  },
  {
   "key": "inkling small",
   "name": "Inkling Small",
   "organization": "Thinking Machines",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 149.1,
   "n_obs": 183,
   "n_configs": 2,
   "configs": [
    {
     "name": "Inkling Small (xhigh)",
     "sci": 152.5,
     "n": 7,
     "thin": true
    },
    {
     "name": "Inkling Small (standard)",
     "sci": 149.1,
     "n": 176,
     "headline": true
    }
   ],
   "release_date": "2026-07-30",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 130
  },
  {
   "key": "qwen35 plus 20260420",
   "name": "Qwen3.5 Plus 2026-04-20",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 149.1,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "o3 pro",
   "name": "o3-pro",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 149.0,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3627b",
   "name": "Qwen3.6 27B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 149.0,
   "n_obs": 42,
   "n_configs": 2,
   "configs": [
    {
     "name": "Qwen3.6 27B (Non-reasoning)",
     "sci": 149.0,
     "n": 42,
     "headline": true
    }
   ],
   "release_date": "2026-04-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 131
  },
  {
   "key": "longcat flash thinking 2601",
   "name": "LongCat-Flash-Thinking-2601",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "reasoning",
   "sci": 148.9,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-01-14",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "step 35 flash 2603",
   "name": "Step 3.5 Flash 2603",
   "organization": "StepFun",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 148.8,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "solar open2250b",
   "name": "Solar Open2 250B",
   "organization": "Upstage",
   "country": "KR",
   "reasoning_status": "hybrid",
   "sci": 148.7,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 3 flash",
   "name": "Gemini 3 Flash",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 148.7,
   "n_obs": 146,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-17",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 132
  },
  {
   "key": "qwen3 max thinking",
   "name": "Qwen3 Max Thinking",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 148.5,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-02-13",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 133
  },
  {
   "key": "qwen35122b a10b",
   "name": "Qwen3.5 122B A10B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 148.3,
   "n_obs": 27,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 134
  },
  {
   "key": "qwen37 flash",
   "name": "Qwen3.7 Flash",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 148.3,
   "n_obs": 7,
   "n_configs": 2,
   "configs": [
    {
     "name": "Qwen3.7 Flash (standard)",
     "sci": 148.3,
     "n": 7,
     "headline": true
    }
   ],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 135
  },
  {
   "key": "solar pro 4",
   "name": "Solar Pro 4",
   "organization": "Upstage",
   "country": "South Korea",
   "reasoning_status": "non_reasoning",
   "sci": 148.2,
   "n_obs": 11,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-06",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 136
  },
  {
   "key": "nova 2 pro",
   "name": "Nova 2 Pro",
   "organization": "Amazon",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 148.2,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "kimi k25",
   "name": "Kimi K2.5",
   "organization": "Moonshot AI",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 148.1,
   "n_obs": 43,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-01-27",
   "cost_usd": 141.11665019999998,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 137
  },
  {
   "key": "glm 45 thinking",
   "name": "GLM-4.5 Thinking",
   "organization": "Zhipu AI",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 148.0,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 138
  },
  {
   "key": "o3",
   "name": "o3",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 147.9,
   "n_obs": 53,
   "n_configs": 5,
   "configs": [
    {
     "name": "o3 (medium)",
     "sci": 147.9,
     "n": 13,
     "headline": true
    },
    {
     "name": "o3 (2025-04-16 snapshot)",
     "sci": 147.8,
     "n": 6
    },
    {
     "name": "o3 (high)",
     "sci": 147.4,
     "n": 17
    },
    {
     "name": "o3 (standard)",
     "sci": 146.6,
     "n": 12
    },
    {
     "name": "o3 (low)",
     "sci": 144.3,
     "n": 5
    }
   ],
   "release_date": "2025-04-16",
   "cost_usd": 1025.331978,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 139
  },
  {
   "key": "longcat 20",
   "name": "LongCat 2.0",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 147.7,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 140
  },
  {
   "key": "gpt 5 mini",
   "name": "GPT-5 mini",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 147.5,
   "n_obs": 127,
   "n_configs": 7,
   "configs": [
    {
     "name": "GPT-5 mini (high)",
     "sci": 147.5,
     "n": 23,
     "headline": true
    },
    {
     "name": "GPT-5 mini (high)",
     "sci": 146.9,
     "n": 14
    },
    {
     "name": "GPT-5 mini (standard)",
     "sci": 146.5,
     "n": 40
    },
    {
     "name": "GPT-5 mini (medium)",
     "sci": 144.6,
     "n": 12
    },
    {
     "name": "GPT-5 mini (low)",
     "sci": 142.0,
     "n": 17
    },
    {
     "name": "GPT-5 mini (minimal)",
     "sci": 137.9,
     "n": 5
    },
    {
     "name": "GPT-5 mini (minimal)",
     "sci": 134.2,
     "n": 16
    }
   ],
   "release_date": "2025-08-07",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 141
  },
  {
   "key": "a x k2",
   "name": "A.X-K2",
   "organization": "SK Telecom",
   "country": "KR",
   "reasoning_status": "unknown",
   "sci": 147.4,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 142
  },
  {
   "key": "kat coder pro v1",
   "name": "KAT-Coder-Pro V1",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 147.4,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 431b it",
   "name": "Gemma 4 31B IT",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 147.4,
   "n_obs": 40,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-04-02",
   "cost_usd": 19.43316902,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 143
  },
  {
   "key": "claude sonnet 45",
   "name": "Claude Sonnet 4.5",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 147.3,
   "n_obs": 67,
   "n_configs": 6,
   "configs": [
    {
     "name": "Claude Sonnet 4.5 (standard)",
     "sci": 147.3,
     "n": 20,
     "headline": true
    },
    {
     "name": "Claude Sonnet 4.5 (standard)",
     "sci": 146.5,
     "n": 8
    },
    {
     "name": "Claude Sonnet 4.5 (20250929 snapshot)",
     "sci": 142.4,
     "n": 39
    }
   ],
   "release_date": "2025-09-29",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 144
  },
  {
   "key": "jt 35b flash",
   "name": "JT-35B-Flash",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 147.3,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "k exaone 236b a23b",
   "name": "K-EXAONE-236B-A23B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 147.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-31",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "grok 4",
   "name": "Grok 4",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 147.2,
   "n_obs": 48,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-07-09",
   "cost_usd": 2881.151911,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 145
  },
  {
   "key": "mai thinking 1",
   "name": "MAI-Thinking-1",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "reasoning",
   "sci": 147.1,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-06-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen35 plus",
   "name": "Qwen3.5 Plus",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 147.1,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 146
  },
  {
   "key": "kat coder pro v25",
   "name": "KAT-Coder-Pro V2.5",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 147.1,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "grok 3 mini",
   "name": "Grok-3 Mini",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 147.0,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-02-17",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "seed 18",
   "name": "Seed 1.8",
   "organization": "ByteDance",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 147.0,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-18",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 147
  },
  {
   "key": "mimo v2 pro",
   "name": "MiMo-V2-Pro",
   "organization": "Xiaomi",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 147.0,
   "n_obs": 25,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-03-18",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 148
  },
  {
   "key": "nemotron 3 ultra 550b a55b",
   "name": "Nemotron 3 Ultra 550B A55B",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 147.0,
   "n_obs": 128,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-06-04",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 149
  },
  {
   "key": "longcat flash thinking",
   "name": "LongCat-Flash-Thinking",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "reasoning",
   "sci": 146.9,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen36 flash",
   "name": "Qwen3.6 Flash",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 146.9,
   "n_obs": 25,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 150
  },
  {
   "key": "deepseek v32 speciale",
   "name": "DeepSeek-V3.2-Speciale",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 146.9,
   "n_obs": 11,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-01",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 151
  },
  {
   "key": "qwen35 omni plus",
   "name": "Qwen3.5 Omni Plus",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 146.8,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "seed 20 mini",
   "name": "Seed-2.0-Mini",
   "organization": "ByteDance",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 146.8,
   "n_obs": 26,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-02-14",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 152
  },
  {
   "key": "deepseek v32 exp thinking",
   "name": "DeepSeek-V3.2-Exp_thinking",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 146.8,
   "n_obs": 28,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 153
  },
  {
   "key": "k2 horizon 7b",
   "name": "K2 Horizon 7B",
   "organization": "MBZUAI",
   "country": "UAE",
   "reasoning_status": "hybrid",
   "sci": 146.8,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-03",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 154
  },
  {
   "key": "minimax m27",
   "name": "MiniMax-M2.7",
   "organization": "MiniMax",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 146.7,
   "n_obs": 133,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 155
  },
  {
   "key": "minimax m21",
   "name": "MiniMax-M2.1",
   "organization": "MiniMax",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 146.6,
   "n_obs": 19,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-23",
   "cost_usd": 114.4574499,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 156
  },
  {
   "key": "hy3 preview",
   "name": "Hy3-preview",
   "organization": "Tencent",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 146.6,
   "n_obs": 29,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-04-23",
   "cost_usd": null,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 157
  },
  {
   "key": "gemini 31 flash lite",
   "name": "Gemini 3.1 Flash-Lite",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 146.6,
   "n_obs": 133,
   "n_configs": 3,
   "configs": [
    {
     "name": "Gemini 3.1 Flash-Lite (high)",
     "sci": 146.6,
     "n": 16,
     "headline": true
    },
    {
     "name": "Gemini 3.1 Flash-Lite (standard)",
     "sci": 142.2,
     "n": 117
    }
   ],
   "release_date": "2026-03-03",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 158
  },
  {
   "key": "grok 4200309",
   "name": "Grok 4.20 0309",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 146.6,
   "n_obs": 49,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 159
  },
  {
   "key": "minimax m25",
   "name": "MiniMax-M2.5",
   "organization": "MiniMax",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 146.5,
   "n_obs": 42,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-02-12",
   "cost_usd": 124.58034959999999,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 160
  },
  {
   "key": "gemma 426b a4b",
   "name": "Gemma 4 26B A4B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 146.4,
   "n_obs": 13,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-04-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 161
  },
  {
   "key": "qwen35 flash",
   "name": "Qwen3.5 Flash",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 146.4,
   "n_obs": 21,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 162
  },
  {
   "key": "o4 mini",
   "name": "o4-mini",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 146.4,
   "n_obs": 45,
   "n_configs": 7,
   "configs": [
    {
     "name": "o4-mini (high)",
     "sci": 146.4,
     "n": 17,
     "headline": true
    },
    {
     "name": "o4-mini (standard)",
     "sci": 143.7,
     "n": 12
    },
    {
     "name": "o4-mini (medium)",
     "sci": 143.6,
     "n": 10
    },
    {
     "name": "o4-mini (low)",
     "sci": 137.9,
     "n": 6
    }
   ],
   "release_date": "2025-04-16",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 163
  },
  {
   "key": "gpt 51 codex mini",
   "name": "GPT-5.1 Codex mini",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 146.3,
   "n_obs": 14,
   "n_configs": 2,
   "configs": [
    {
     "name": "GPT-5.1 Codex mini (standard)",
     "sci": 146.3,
     "n": 14,
     "headline": true
    }
   ],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 164
  },
  {
   "key": "gemini 25 pro exp 0325",
   "name": "Gemini 2.5 Pro Exp 03.25",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 146.3,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-03-25",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 165
  },
  {
   "key": "kimi k2 thinking",
   "name": "Kimi K2 Thinking",
   "organization": "Moonshot AI",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 146.2,
   "n_obs": 42,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-05",
   "cost_usd": 308.1504595,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 166
  },
  {
   "key": "glm 47",
   "name": "GLM-4.7",
   "organization": "Zhipu AI",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 146.2,
   "n_obs": 42,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-22",
   "cost_usd": 147.1413482,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 167
  },
  {
   "key": "gpt 55 instant (may 2026)",
   "name": "GPT-5.5 Instant (May 2026)",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 146.1,
   "n_obs": 14,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 168
  },
  {
   "key": "nova 20 pro preview",
   "name": "Nova 2.0 Pro Preview",
   "organization": "Amazon",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 146.0,
   "n_obs": 16,
   "n_configs": 3,
   "configs": [
    {
     "name": "Nova 2.0 Pro Preview (medium)",
     "sci": 146.0,
     "n": 5,
     "headline": true
    },
    {
     "name": "Nova 2.0 Pro Preview (low)",
     "sci": 143.4,
     "n": 5
    },
    {
     "name": "Nova 2.0 Pro Preview (Non-reasoning)",
     "sci": 135.2,
     "n": 6
    }
   ],
   "release_date": "2025-12-02",
   "cost_usd": null,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 169
  },
  {
   "key": "ling 30 flash",
   "name": "Ling 3.0 Flash",
   "organization": "Ant Group",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 146.0,
   "n_obs": 14,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-04",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 170
  },
  {
   "key": "sarvam 105b",
   "name": "Sarvam-105B",
   "organization": "Sarvam AI",
   "country": "India",
   "reasoning_status": "unknown",
   "sci": 146.0,
   "n_obs": 3,
   "n_configs": 2,
   "configs": [
    {
     "name": "Sarvam-105B (standard)",
     "sci": 146.0,
     "n": 0,
     "headline": true
    },
    {
     "name": "Sarvam-105B (high)",
     "sci": 134.1,
     "n": 3
    }
   ],
   "release_date": "2026-03-06",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mimo v2 omni",
   "name": "MiMo-V2-Omni",
   "organization": "Xiaomi",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 146.0,
   "n_obs": 8,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-03-18",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 171
  },
  {
   "key": "k exaone 200803",
   "name": "K-EXAONE 2.0 0803",
   "organization": "LG AI Research",
   "country": "KR",
   "reasoning_status": "hybrid",
   "sci": 145.9,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek reasoner",
   "name": "Deepseek Reasoner",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 145.6,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 172
  },
  {
   "key": "gemini 25 pro preview (mar' 25)",
   "name": "Gemini 2.5 Pro Preview (Mar' 25)",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 145.6,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "grok 420 beta reasoning",
   "name": "Grok-4.20 Beta Reasoning",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 145.6,
   "n_obs": 121,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 173
  },
  {
   "key": "nemotron 3 super fp8",
   "name": "Nemotron-3-Super FP8",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 145.5,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-03-11",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "nemotron 3 super nvfp4",
   "name": "Nemotron-3-Super NVFP4",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 145.4,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-03-11",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 25 pro preview 0325",
   "name": "Gemini 2.5 Pro Preview 03.25",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 145.4,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 174
  },
  {
   "key": "g9v339a5b",
   "name": "G9v3-39A5B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 145.3,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "step 35 flash",
   "name": "Step-3.5-Flash",
   "organization": "StepFun",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 145.3,
   "n_obs": 14,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-02-02",
   "cost_usd": 74.80080609999999,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 175
  },
  {
   "key": "grok 4 fast",
   "name": "Grok 4 Fast",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 145.2,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-08-28",
   "cost_usd": 17.448047600000002,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 176
  },
  {
   "key": "qwen3535b a3b",
   "name": "Qwen3.5 35B A3B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 145.2,
   "n_obs": 29,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 177
  },
  {
   "key": "minimax m2",
   "name": "MiniMax M2",
   "organization": "MiniMax",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 145.2,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-10-27",
   "cost_usd": 115.7639499,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 178
  },
  {
   "key": "gemini 25 pro preview (may' 25)",
   "name": "Gemini 2.5 Pro Preview (May' 25)",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 145.1,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral medium 35",
   "name": "Mistral Medium 3.5",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 145.0,
   "n_obs": 104,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-04-29",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 179
  },
  {
   "key": "grok 3 mini reasoning",
   "name": "Grok 3 mini Reasoning",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 145.0,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "ling 261t",
   "name": "Ling-2.6-1T",
   "organization": "Ant Group",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 144.9,
   "n_obs": 11,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 180
  },
  {
   "key": "qwen3635b a3b",
   "name": "Qwen3.6 35B A3B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 144.9,
   "n_obs": 31,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-04-16",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 181
  },
  {
   "key": "glm 46",
   "name": "GLM-4.6",
   "organization": "Zhipu AI",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 144.7,
   "n_obs": 33,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-30",
   "cost_usd": 111.1159132,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 182
  },
  {
   "key": "mimo v2 flash",
   "name": "MiMo-V2-Flash",
   "organization": "Xiaomi",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 144.7,
   "n_obs": 16,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-16",
   "cost_usd": 21.3815107,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 183
  },
  {
   "key": "deepseek v31 terminus",
   "name": "DeepSeek V3.1 Terminus",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 144.6,
   "n_obs": 8,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 184
  },
  {
   "key": "k exaone",
   "name": "K-EXAONE",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 144.6,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 25 pro preview 0605",
   "name": "Gemini 2.5 Pro Preview 06-05",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 144.5,
   "n_obs": 11,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-06-05",
   "cost_usd": null,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 185
  },
  {
   "key": "mercury 25 preview",
   "name": "Mercury 2.5 Preview",
   "organization": "Inception",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 144.4,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-01",
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "nova 20 lite",
   "name": "Nova 2.0 Lite",
   "organization": "Amazon",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 144.4,
   "n_obs": 13,
   "n_configs": 4,
   "configs": [
    {
     "name": "Nova 2.0 Lite (medium)",
     "sci": 144.4,
     "n": 3,
     "headline": true
    },
    {
     "name": "Nova 2.0 Lite (high)",
     "sci": 143.7,
     "n": 4
    },
    {
     "name": "Nova 2.0 Lite (low)",
     "sci": 138.3,
     "n": 3
    },
    {
     "name": "Nova 2.0 Lite (Non-reasoning)",
     "sci": 132.7,
     "n": 3
    }
   ],
   "release_date": "2025-12-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "doubao seed code",
   "name": "Doubao Seed Code",
   "organization": "ByteDance",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 144.2,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 25 pro",
   "name": "Gemini 2.5 Pro",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 144.2,
   "n_obs": 42,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-05-20",
   "cost_usd": 648.3154525,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 186
  },
  {
   "key": "qwen3 max thinking (preview)",
   "name": "Qwen3 Max Thinking (Preview)",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 143.9,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "nova 20 omni",
   "name": "Nova 2.0 Omni",
   "organization": "Amazon",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 143.9,
   "n_obs": 9,
   "n_configs": 3,
   "configs": [
    {
     "name": "Nova 2.0 Omni (medium)",
     "sci": 143.9,
     "n": 3,
     "headline": true
    },
    {
     "name": "Nova 2.0 Omni (low)",
     "sci": 139.7,
     "n": 3
    },
    {
     "name": "Nova 2.0 Omni (Non-reasoning)",
     "sci": 131.7,
     "n": 3
    }
   ],
   "release_date": "2025-12-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 vl 235b a22b thinking",
   "name": "Qwen3 VL 235B A22B Thinking",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 143.7,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 187
  },
  {
   "key": "qwen3 max",
   "name": "Qwen3 Max",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 143.7,
   "n_obs": 19,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-23",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 188
  },
  {
   "key": "claude 4 sonnet",
   "name": "Claude 4 Sonnet",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 143.7,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-05-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 189
  },
  {
   "key": "apriel v1615b thinker",
   "name": "Apriel-v1.6-15B-Thinker",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 143.6,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "trinity large thinking",
   "name": "Trinity Large Thinking",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "reasoning",
   "sci": 143.6,
   "n_obs": 15,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 190
  },
  {
   "key": "gpt oss 120b high",
   "name": "GPT OSS 120B High",
   "organization": "OpenAI",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 143.6,
   "n_obs": 16,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-08-05",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 191
  },
  {
   "key": "qwen3235b a22b thinking 2507",
   "name": "Qwen3-235B-A22B-Thinking-2507",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 143.5,
   "n_obs": 27,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-07-25",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 192
  },
  {
   "key": "claude haiku 45",
   "name": "Claude Haiku 4.5",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 143.5,
   "n_obs": 49,
   "n_configs": 3,
   "configs": [
    {
     "name": "Claude Haiku 4.5 (standard)",
     "sci": 143.5,
     "n": 14,
     "headline": true
    },
    {
     "name": "Claude Haiku 4.5 (standard)",
     "sci": 143.2,
     "n": 7
    },
    {
     "name": "Claude Haiku 4.5 (20251001 snapshot)",
     "sci": 136.6,
     "n": 28
    }
   ],
   "release_date": "2025-10-15",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 193
  },
  {
   "key": "exaone 4533b",
   "name": "EXAONE 4.5 33B",
   "organization": "LG AI Research",
   "country": "South Korea",
   "reasoning_status": "non_reasoning",
   "sci": 143.5,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-04-09",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "glm 5 turbo",
   "name": "GLM-5-Turbo",
   "organization": "Zhipu AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 143.5,
   "n_obs": 22,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 194
  },
  {
   "key": "nova 2 lite",
   "name": "Nova 2 Lite",
   "organization": "Amazon",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 143.2,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek v32",
   "name": "DeepSeek V3.2",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 143.1,
   "n_obs": 43,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-01",
   "cost_usd": 39.05466593,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 195
  },
  {
   "key": "qwen3235b a22b 2507",
   "name": "Qwen3 235B A22B 2507",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 143.0,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 196
  },
  {
   "key": "qwen354b",
   "name": "Qwen3.5 4B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 142.9,
   "n_obs": 8,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-03-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 197
  },
  {
   "key": "laguna xs 21",
   "name": "Laguna XS 2.1",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 142.9,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen35 omni flash",
   "name": "Qwen3.5 Omni Flash",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 142.8,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "o1 pro",
   "name": "o1-pro",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 142.7,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-12-17",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 25 flash",
   "name": "Gemini 2.5 Flash",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 142.6,
   "n_obs": 25,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-05-20",
   "cost_usd": 84.27270970000001,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 198
  },
  {
   "key": "deepseek v31 thinking",
   "name": "DeepSeek-V3.1_thinking",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 142.6,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "k2 horizon 37b",
   "name": "K2 Horizon 3.7B",
   "organization": "MBZUAI",
   "country": "UAE",
   "reasoning_status": "hybrid",
   "sci": 142.5,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-03",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 199
  },
  {
   "key": "qwen3 vl 235b a22b",
   "name": "Qwen3 VL 235B A22B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 142.5,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "grok 41 fast",
   "name": "Grok 4.1 Fast",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 142.4,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 200
  },
  {
   "key": "mercury 2",
   "name": "Mercury 2",
   "organization": "Inception",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 142.4,
   "n_obs": 12,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 201
  },
  {
   "key": "gemini 25 flash preview (sep '25)",
   "name": "Gemini 2.5 Flash Preview (Sep '25)",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 142.3,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "glm 45",
   "name": "GLM-4.5",
   "organization": "Zhipu AI",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 142.3,
   "n_obs": 11,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-07-28",
   "cost_usd": 408.49530319999997,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 202
  },
  {
   "key": "mistral medium 2604",
   "name": "Mistral Medium 2604",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 142.2,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 203
  },
  {
   "key": "o3 mini",
   "name": "o3-mini",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 142.2,
   "n_obs": 43,
   "n_configs": 5,
   "configs": [
    {
     "name": "o3-mini (high)",
     "sci": 142.2,
     "n": 15,
     "headline": true
    },
    {
     "name": "o3-mini (high)",
     "sci": 140.1,
     "n": 8
    },
    {
     "name": "o3-mini (standard)",
     "sci": 140.1,
     "n": 7
    },
    {
     "name": "o3-mini (medium)",
     "sci": 139.5,
     "n": 7
    },
    {
     "name": "o3-mini (low)",
     "sci": 135.7,
     "n": 6
    }
   ],
   "release_date": "2025-01-30",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 204
  },
  {
   "key": "qwen3 next 80b a3b thinking",
   "name": "Qwen3-Next-80B-A3B-Thinking",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 142.2,
   "n_obs": 15,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-10",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 205
  },
  {
   "key": "ernie 50 thinking preview",
   "name": "ERNIE 5.0 Thinking Preview",
   "organization": "Baidu",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 142.0,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "sarvam 30b",
   "name": "Sarvam-30B",
   "organization": "Sarvam AI",
   "country": "India",
   "reasoning_status": "unknown",
   "sci": 142.0,
   "n_obs": 3,
   "n_configs": 2,
   "configs": [
    {
     "name": "Sarvam-30B (standard)",
     "sci": 142.0,
     "n": 0,
     "headline": true
    },
    {
     "name": "Sarvam-30B (high)",
     "sci": 130.1,
     "n": 3
    }
   ],
   "release_date": "2026-03-06",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "laguna m 1",
   "name": "Laguna M.1",
   "organization": "Poolside",
   "country": "US",
   "reasoning_status": "unknown",
   "sci": 142.0,
   "n_obs": 79,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 206
  },
  {
   "key": "seed oss 36b instruct",
   "name": "Seed-OSS-36B-Instruct",
   "organization": "ByteDance",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 141.7,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek r10528 (may '25)",
   "name": "DeepSeek R1 0528 (May '25)",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 141.4,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "grok 3 mini beta",
   "name": "Grok 3 Mini Beta",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 141.3,
   "n_obs": 7,
   "n_configs": 2,
   "configs": [
    {
     "name": "Grok 3 Mini Beta (high)",
     "sci": 141.3,
     "n": 3,
     "headline": true
    },
    {
     "name": "Grok 3 Mini Beta (low)",
     "sci": 137.0,
     "n": 4
    }
   ],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "solar pro 3",
   "name": "Solar Pro 3",
   "organization": "Upstage",
   "country": "South Korea",
   "reasoning_status": "non_reasoning",
   "sci": 141.3,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 207
  },
  {
   "key": "minicpm sala",
   "name": "MiniCPM-SALA",
   "organization": "OpenBMB",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 141.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-02-11",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "granite 4230b",
   "name": "Granite 4.2 30B",
   "organization": "IBM",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 141.3,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-25",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 208
  },
  {
   "key": "kat coder air v25",
   "name": "KAT-Coder-Air V2.5",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 141.2,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mercury 25",
   "name": "Mercury 2.5",
   "organization": "Inception",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 141.2,
   "n_obs": 108,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-08",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 209
  },
  {
   "key": "gpt 5 nano",
   "name": "GPT-5 nano",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 141.2,
   "n_obs": 91,
   "n_configs": 8,
   "configs": [
    {
     "name": "GPT-5 nano (high)",
     "sci": 141.2,
     "n": 16,
     "headline": true
    },
    {
     "name": "GPT-5 nano (high)",
     "sci": 140.3,
     "n": 14
    },
    {
     "name": "GPT-5 nano (standard)",
     "sci": 140.1,
     "n": 30
    },
    {
     "name": "GPT-5 nano (medium)",
     "sci": 136.3,
     "n": 7
    },
    {
     "name": "GPT-5 nano (low)",
     "sci": 133.2,
     "n": 19
    },
    {
     "name": "GPT-5 nano (minimal)",
     "sci": 131.0,
     "n": 5
    }
   ],
   "release_date": "2025-08-07",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 210
  },
  {
   "key": "nemotron cascade 230b a3b",
   "name": "Nemotron Cascade 2 30B A3B",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 141.2,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-03-19",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 vl 32b thinking",
   "name": "Qwen3 VL 32B Thinking",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 141.1,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek r10528",
   "name": "DeepSeek-R1-0528",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 141.1,
   "n_obs": 13,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-05-28",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 211
  },
  {
   "key": "deepseek r1 zero",
   "name": "DeepSeek R1 Zero",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 141.1,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-20",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "hypernova 60b 2605",
   "name": "HyperNova 60B 2605",
   "organization": "Multiverse Computing",
   "country": "Spain",
   "reasoning_status": "reasoning",
   "sci": 141.1,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-05-06",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 212
  },
  {
   "key": "qwen3 vl 32b",
   "name": "Qwen3 VL 32B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 141.0,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "apriel v1515b thinker",
   "name": "Apriel-v1.5-15B-Thinker",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 141.0,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 next 80b a3b",
   "name": "Qwen3 Next 80B A3B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 140.9,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 213
  },
  {
   "key": "gpt oss 20b high",
   "name": "GPT OSS 20B High",
   "organization": "OpenAI",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 140.9,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-08-05",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 214
  },
  {
   "key": "glm 45 air",
   "name": "GLM-4.5-Air",
   "organization": "Zhipu AI",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 140.8,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-07-28",
   "cost_usd": 93.37700393,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 215
  },
  {
   "key": "claude opus 4",
   "name": "Claude Opus 4",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 140.7,
   "n_obs": 24,
   "n_configs": 3,
   "configs": [
    {
     "name": "Claude Opus 4 (standard)",
     "sci": 142.6,
     "n": 5,
     "thin": true
    },
    {
     "name": "Claude Opus 4 (standard)",
     "sci": 140.7,
     "n": 19,
     "headline": true
    }
   ],
   "release_date": "2025-05-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 216
  },
  {
   "key": "gemma 412b",
   "name": "Gemma 4 12B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 140.7,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 217
  },
  {
   "key": "gpt oss 120b",
   "name": "gpt-oss-120B",
   "organization": "OpenAI",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 140.7,
   "n_obs": 39,
   "n_configs": 2,
   "configs": [
    {
     "name": "gpt-oss-120B (standard)",
     "sci": 140.7,
     "n": 33,
     "headline": true
    },
    {
     "name": "gpt-oss-120B (low)",
     "sci": 138.4,
     "n": 6
    }
   ],
   "release_date": "2025-08-05",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 218
  },
  {
   "key": "o1",
   "name": "o1",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 140.6,
   "n_obs": 25,
   "n_configs": 4,
   "configs": [
    {
     "name": "o1 (high)",
     "sci": 142.5,
     "n": 8,
     "thin": true
    },
    {
     "name": "o1 (standard)",
     "sci": 140.6,
     "n": 11,
     "headline": true
    },
    {
     "name": "o1 (medium)",
     "sci": 140.4,
     "n": 6
    }
   ],
   "release_date": "2024-12-05",
   "cost_usd": 3270.35733,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 219
  },
  {
   "key": "magistral medium 12",
   "name": "Magistral Medium 1.2",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 140.6,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 220
  },
  {
   "key": "kimi k20905",
   "name": "Kimi K2 0905",
   "organization": "Moonshot AI",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 140.5,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-05",
   "cost_usd": 126.5650654,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 coder next",
   "name": "Qwen3 Coder Next",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 140.5,
   "n_obs": 13,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 221
  },
  {
   "key": "jt mini",
   "name": "JT-MINI",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 140.4,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "nemotron 3 super",
   "name": "Nemotron-3-Super",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 140.4,
   "n_obs": 28,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-03-11",
   "cost_usd": 140.39501744999998,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 222
  },
  {
   "key": "north mini code",
   "name": "North Mini Code",
   "organization": "Cohere",
   "country": "Canada",
   "reasoning_status": "unknown",
   "sci": 140.4,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-06-09",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 223
  },
  {
   "key": "falcon h1r 7b",
   "name": "Falcon-H1R-7B",
   "organization": "TII",
   "country": "UAE",
   "reasoning_status": "non_reasoning",
   "sci": 140.4,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 25 flash preview 0520",
   "name": "Gemini 2.5 Flash Preview 05.20",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 140.3,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 25 flash preview 092025",
   "name": "Gemini 2.5 Flash Preview 09.2025",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 140.2,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 224
  },
  {
   "key": "gpt oss 20b",
   "name": "GPT OSS 20B",
   "organization": "OpenAI",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 140.1,
   "n_obs": 11,
   "n_configs": 3,
   "configs": [
    {
     "name": "GPT OSS 20B (standard)",
     "sci": 140.1,
     "n": 6,
     "headline": true
    },
    {
     "name": "GPT OSS 20B (low)",
     "sci": 136.2,
     "n": 5
    }
   ],
   "release_date": "2025-08-05",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 225
  },
  {
   "key": "nemotron 3 nano (30b a3b)",
   "name": "Nemotron 3 Nano (30B A3B)",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 140.1,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-15",
   "cost_usd": 13.732894550000001,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 226
  },
  {
   "key": "minicpm52b",
   "name": "MiniCPM5-2B",
   "organization": "OpenBMB",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 140.1,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-07",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 227
  },
  {
   "key": "deepseek v31",
   "name": "DeepSeek-V3.1",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 140.0,
   "n_obs": 12,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-10",
   "cost_usd": 99.60698752500002,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 228
  },
  {
   "key": "grok 41 thinking",
   "name": "Grok-4.1 Thinking",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 140.0,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-11-17",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 229
  },
  {
   "key": "longcat flash chat",
   "name": "LongCat-Flash-Chat",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 139.9,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-08-29",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen359b",
   "name": "Qwen3.5 9B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 139.9,
   "n_obs": 20,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-03-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 230
  },
  {
   "key": "mi:dm k 25 pro",
   "name": "Mi:dm K 2.5 Pro",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 139.9,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3235b a22b",
   "name": "Qwen3 235B A22B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 139.7,
   "n_obs": 12,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-04-29",
   "cost_usd": 103.67145945,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 231
  },
  {
   "key": "ring 1t",
   "name": "Ring-1T",
   "organization": "Ant Group",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 139.6,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "kimi k2",
   "name": "Kimi K2",
   "organization": "Moonshot AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 139.5,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-07-11",
   "cost_usd": 180.802504215,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "phi 4 reasoning plus",
   "name": "Phi 4 Reasoning Plus",
   "organization": "Microsoft",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 139.5,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-04-30",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3235b a22b instruct 2507",
   "name": "Qwen3-235B-A22B-Instruct-2507",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 139.4,
   "n_obs": 21,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-07-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 232
  },
  {
   "key": "longcat flash lite",
   "name": "LongCat-Flash-Lite",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 139.4,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-02-05",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "minimax m180k",
   "name": "MiniMax M1 80K",
   "organization": "MiniMax",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 139.3,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-06-16",
   "cost_usd": 114.97301530000001,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "intellect 3",
   "name": "INTELLECT-3",
   "organization": "Prime Intellect",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 139.3,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-11-26",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "glm 47 flash",
   "name": "GLM-4.7-Flash",
   "organization": "Zhipu AI",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 139.3,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-01-19",
   "cost_usd": 22.190865860000002,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 233
  },
  {
   "key": "tri 21b think preview",
   "name": "Tri-21B-think Preview",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 139.2,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "claude opus 41",
   "name": "Claude Opus 4.1",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 139.1,
   "n_obs": 27,
   "n_configs": 5,
   "configs": [
    {
     "name": "Claude Opus 4.1 (standard)",
     "sci": 145.9,
     "n": 7,
     "thin": true
    },
    {
     "name": "Claude Opus 4.1 (20250805 snapshot)",
     "sci": 139.1,
     "n": 20,
     "headline": true
    }
   ],
   "release_date": "2025-08-05",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 234
  },
  {
   "key": "cogito v21",
   "name": "Cogito v2.1",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "reasoning",
   "sci": 139.1,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek chat",
   "name": "Deepseek Chat",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 139.1,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 235
  },
  {
   "key": "ling 26 flash",
   "name": "Ling 2.6 Flash",
   "organization": "Ant Group",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 139.0,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 236
  },
  {
   "key": "qwen 3 max",
   "name": "Qwen 3 Max",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 139.0,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": 257.846688825,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 237
  },
  {
   "key": "grok code fast 1",
   "name": "Grok Code Fast 1",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 139.0,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 238
  },
  {
   "key": "deepseek r1",
   "name": "DeepSeek R1",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 139.0,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-20",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 239
  },
  {
   "key": "laguna xs 2",
   "name": "Laguna XS.2",
   "organization": "Poolside",
   "country": "US",
   "reasoning_status": "unknown",
   "sci": 138.9,
   "n_obs": 78,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 240
  },
  {
   "key": "gemini 25 flash lite preview (sep '25)",
   "name": "Gemini 2.5 Flash-Lite Preview (Sep '25)",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 138.9,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "granite 428b",
   "name": "Granite 4.2 8B",
   "organization": "IBM",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 138.9,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-25",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 241
  },
  {
   "key": "deepseek v2",
   "name": "DeepSeek-V2",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 138.9,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen330b a3b 2507",
   "name": "Qwen3 30B A3B 2507",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 138.7,
   "n_obs": 12,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-07-30",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 242
  },
  {
   "key": "llama nemotron super 49b v15",
   "name": "Llama Nemotron Super 49B v1.5",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 138.7,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-07-25",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "kimi k15",
   "name": "Kimi-k1.5",
   "organization": "Moonshot AI",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 138.6,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-20",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 vl 30b a3b thinking",
   "name": "Qwen3 VL 30B A3B Thinking",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 138.5,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 vl 235b a22b instruct",
   "name": "Qwen3 VL 235B A22B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 138.4,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-22",
   "cost_usd": 71.807963175,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 243
  },
  {
   "key": "qwen3 next 80b a3b instruct",
   "name": "Qwen3-Next-80B-A3B-Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 138.4,
   "n_obs": 17,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-10",
   "cost_usd": 132.972724,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 244
  },
  {
   "key": "kimi k2 instruct",
   "name": "Kimi-K2-Instruct",
   "organization": "Moonshot AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 138.3,
   "n_obs": 26,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-07-11",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 245
  },
  {
   "key": "mistral small 4",
   "name": "Mistral Small 4",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 138.3,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 246
  },
  {
   "key": "tri 21b think",
   "name": "Tri-21B-Think",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 138.2,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "nemotron 35 lightning (30b a3b)",
   "name": "Nemotron 3.5 Lightning (30B A3B)",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 138.2,
   "n_obs": 98,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-11",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 247
  },
  {
   "key": "laguna s 21",
   "name": "Laguna S 2.1",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 138.1,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 248
  },
  {
   "key": "hyperclova x seed think (32b)",
   "name": "HyperCLOVA X SEED Think (32B)",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 138.0,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "motif 2127b reasoning",
   "name": "Motif-2-12.7B-Reasoning",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "reasoning",
   "sci": 138.0,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "k2 v2",
   "name": "K2-V2",
   "organization": "MBZUAI",
   "country": "UAE",
   "reasoning_status": "unknown",
   "sci": 138.0,
   "n_obs": 9,
   "n_configs": 3,
   "configs": [
    {
     "name": "K2-V2 (high)",
     "sci": 138.0,
     "n": 3,
     "headline": true
    },
    {
     "name": "K2-V2 (medium)",
     "sci": 135.0,
     "n": 3
    },
    {
     "name": "K2-V2 (low)",
     "sci": 130.3,
     "n": 3
    }
   ],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 vl 30b a3b",
   "name": "Qwen3 VL 30B A3B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 138.0,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "magistral medium 2509",
   "name": "Magistral Medium 2509",
   "organization": "Mistral",
   "country": "FR",
   "reasoning_status": "reasoning",
   "sci": 137.9,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 249
  },
  {
   "key": "k2 think v2",
   "name": "K2 Think V2",
   "organization": "MBZUAI",
   "country": "UAE",
   "reasoning_status": "unknown",
   "sci": 137.8,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "claude sonnet 4",
   "name": "Claude Sonnet 4",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 137.7,
   "n_obs": 29,
   "n_configs": 2,
   "configs": [
    {
     "name": "Claude Sonnet 4 (20250514 snapshot)",
     "sci": 137.7,
     "n": 29,
     "headline": true
    }
   ],
   "release_date": "2025-05-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 250
  },
  {
   "key": "exaone 4032b",
   "name": "EXAONE 4.0 32B",
   "organization": "LG AI Research",
   "country": "South Korea",
   "reasoning_status": "non_reasoning",
   "sci": 137.7,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mi:dm k 25 pro preview",
   "name": "Mi:dm K 2.5 Pro Preview",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 137.7,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "ling 1t",
   "name": "Ling-1T",
   "organization": "Ant Group",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 137.6,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "kimi k2 instruct 0905",
   "name": "Kimi K2-Instruct-0905",
   "organization": "Moonshot AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 137.6,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-05",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 251
  },
  {
   "key": "phi 4 reasoning",
   "name": "Phi 4 Reasoning",
   "organization": "Microsoft",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 137.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-04-30",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen34b 2507",
   "name": "Qwen3 4B 2507",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 137.4,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 vl 8b thinking",
   "name": "Qwen3 VL 8B Thinking",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 137.4,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "grok 41 fast non reasoning",
   "name": "Grok 4.1 Fast Non Reasoning",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 137.4,
   "n_obs": 33,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 252
  },
  {
   "key": "deepseek r1 (jan '25)",
   "name": "DeepSeek R1 (Jan '25)",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 137.3,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 253
  },
  {
   "key": "o1 preview",
   "name": "o1-preview",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 137.1,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-09-12",
   "cost_usd": null,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 254
  },
  {
   "key": "gpt 45",
   "name": "GPT-4.5",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 137.1,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-02-27",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 31 nemotron nano 8b v1",
   "name": "Llama 3.1 Nemotron Nano 8B V1",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 137.0,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-03-18",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "solar open 100b",
   "name": "Solar Open 100B",
   "organization": "Upstage",
   "country": "South Korea",
   "reasoning_status": "reasoning",
   "sci": 137.0,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen332b",
   "name": "Qwen3 32B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 136.9,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-04-29",
   "cost_usd": 35.07512203,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 255
  },
  {
   "key": "gpt 45 preview",
   "name": "GPT-4.5 Preview",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 136.9,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-02-27",
   "cost_usd": null,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 256
  },
  {
   "key": "minimax m140k",
   "name": "MiniMax M1 40K",
   "organization": "MiniMax",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 136.9,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-06-16",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 coder 480b a35b instruct",
   "name": "Qwen3-Coder 480B A35B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 136.8,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-31",
   "cost_usd": 158.5285791,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 257
  },
  {
   "key": "grok 3",
   "name": "Grok-3",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 136.8,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-02-17",
   "cost_usd": 415.607208,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 258
  },
  {
   "key": "gemini 25 flash preview 0417",
   "name": "Gemini 2.5 Flash Preview 04.17",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 136.8,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "magistral small 12",
   "name": "Magistral Small 1.2",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 136.7,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 259
  },
  {
   "key": "magistral small 1",
   "name": "Magistral Small 1",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 136.7,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen330b a3b",
   "name": "Qwen3 30B A3B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 136.7,
   "n_obs": 20,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-04-29",
   "cost_usd": 10.777682290000001,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 260
  },
  {
   "key": "qwen3 vl 32b instruct",
   "name": "Qwen3 VL 32B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 136.5,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-22",
   "cost_usd": 275.3600717,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 261
  },
  {
   "key": "hermes 4 llama 31405b",
   "name": "Hermes 4 - Llama-3.1 405B",
   "organization": "Nous Research",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 136.2,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "diffusiongemma 26b a4b",
   "name": "DiffusionGemma 26B A4B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 136.2,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-06-10",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 20 flash thinking",
   "name": "Gemini 2.0 Flash Thinking",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 136.2,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-21",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "nvidia nemotron nano 12b v2 vl",
   "name": "NVIDIA Nemotron Nano 12B v2 VL",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 136.2,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-10-28",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen314b",
   "name": "Qwen3 14B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 136.1,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-04-29",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 262
  },
  {
   "key": "llama 31 nemotron ultra 253b v1",
   "name": "Llama 3.1 Nemotron Ultra 253B v1",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 136.1,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-04-07",
   "cost_usd": 486.8198586,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "devstral 2512",
   "name": "Devstral 2512",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 136.0,
   "n_obs": 19,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 263
  },
  {
   "key": "ling flash 20",
   "name": "Ling-flash-2.0",
   "organization": "Ant Group",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 135.9,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "celeris 1",
   "name": "Celeris-1",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 135.8,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 264
  },
  {
   "key": "deepseek v30324",
   "name": "DeepSeek-V3-0324",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 135.7,
   "n_obs": 16,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 265
  },
  {
   "key": "qwen330b a3b 2507 instruct",
   "name": "Qwen3 30B A3B 2507 Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 135.7,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 266
  },
  {
   "key": "nvidia nemotron nano 9b v2",
   "name": "NVIDIA Nemotron Nano 9B v2",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 135.6,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-08-18",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek r1 distill qwen 32b",
   "name": "DeepSeek R1 Distill Qwen 32B",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 135.6,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-20",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 omni 30b a3b",
   "name": "Qwen3 Omni 30B A3B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 135.6,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 41",
   "name": "GPT-4.1",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 135.6,
   "n_obs": 27,
   "n_configs": 3,
   "configs": [
    {
     "name": "GPT-4.1 (2025-04-14 snapshot)",
     "sci": 135.6,
     "n": 20,
     "headline": true
    },
    {
     "name": "GPT-4.1 (standard)",
     "sci": 135.4,
     "n": 7
    }
   ],
   "release_date": "2025-04-14",
   "cost_usd": 277.806692,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 267
  },
  {
   "key": "solar pro 2",
   "name": "Solar Pro 2",
   "organization": "Upstage",
   "country": "South Korea",
   "reasoning_status": "non_reasoning",
   "sci": 135.5,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "magistral small 2509",
   "name": "Magistral Small 2509",
   "organization": "Mistral",
   "country": "FR",
   "reasoning_status": "reasoning",
   "sci": 135.4,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "solar pro 2 (preview)",
   "name": "Solar Pro 2 (Preview)",
   "organization": "Upstage",
   "country": "South Korea",
   "reasoning_status": "non_reasoning",
   "sci": 135.4,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "glm 46v",
   "name": "GLM-4.6V",
   "organization": "Zhipu AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 135.4,
   "n_obs": 16,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-08",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 268
  },
  {
   "key": "qwen3 vl 30b a3b instruct",
   "name": "Qwen3 VL 30B A3B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 135.3,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-22",
   "cost_usd": 29.29074836366,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 269
  },
  {
   "key": "hermes 4 llama 3170b",
   "name": "Hermes 4 - Llama-3.1 70B",
   "organization": "Nous Research",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 135.3,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "ling 30 tiny",
   "name": "Ling 3.0 Tiny",
   "organization": "Ant Group",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 135.2,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 270
  },
  {
   "key": "granite 338b instruct",
   "name": "Granite 3.3 8B Instruct",
   "organization": "IBM",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 135.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-04-16",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral small 2603",
   "name": "Mistral Small 2603",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 135.1,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 271
  },
  {
   "key": "gemini 25 flash lite",
   "name": "Gemini 2.5 Flash-Lite",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 135.1,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-06-17",
   "cost_usd": 29.8749326,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 272
  },
  {
   "key": "qwq 32b",
   "name": "QwQ-32B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "reasoning",
   "sci": 135.1,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-03-05",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral medium 2508",
   "name": "Mistral Medium 2508",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 135.0,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 vl 4b thinking",
   "name": "Qwen3 VL 4B Thinking",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 135.0,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek r1 distill llama 70b",
   "name": "DeepSeek R1 Distill Llama 70B",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 134.9,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-20",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "o1 mini",
   "name": "o1-mini",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 134.8,
   "n_obs": 6,
   "n_configs": 3,
   "configs": [
    {
     "name": "o1-mini (medium)",
     "sci": 134.8,
     "n": 6,
     "headline": true
    }
   ],
   "release_date": "2024-09-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 273
  },
  {
   "key": "glm 45v",
   "name": "GLM-4.5V",
   "organization": "Zhipu AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 134.8,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-08-11",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 5 (chatgpt)",
   "name": "GPT-5 (ChatGPT)",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 134.6,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "magistral small 2506",
   "name": "Magistral Small 2506",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 134.6,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-06-10",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek r1 distill qwen 14b",
   "name": "DeepSeek R1 Distill Qwen 14B",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 134.5,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-20",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "olmo 3132b think",
   "name": "Olmo 3.1 32B Think",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 134.5,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 33 nemotron super 49b v1 thinking",
   "name": "Llama 3.3 Nemotron Super 49B v1 Thinking",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 134.4,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-03-18",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 41 mini",
   "name": "GPT-4.1 mini",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 134.4,
   "n_obs": 19,
   "n_configs": 2,
   "configs": [
    {
     "name": "GPT-4.1 mini (2025-04-14 snapshot)",
     "sci": 134.4,
     "n": 14,
     "headline": true
    },
    {
     "name": "GPT-4.1 mini (standard)",
     "sci": 134.2,
     "n": 5
    }
   ],
   "release_date": "2025-04-14",
   "cost_usd": 53.7334464,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 274
  },
  {
   "key": "claude 37 sonnet",
   "name": "Claude 3.7 Sonnet",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 134.3,
   "n_obs": 23,
   "n_configs": 6,
   "configs": [
    {
     "name": "Claude 3.7 Sonnet (standard)",
     "sci": 136.6,
     "n": 7,
     "thin": true
    },
    {
     "name": "Claude 3.7 Sonnet (20250219 snapshot)",
     "sci": 134.3,
     "n": 16,
     "headline": true
    }
   ],
   "release_date": "2025-02-24",
   "cost_usd": 524.7573,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 275
  },
  {
   "key": "mistral small 324b instruct",
   "name": "Mistral Small 3 24B Instruct",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 134.3,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-30",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 25 flash lite preview 092025",
   "name": "Gemini 2.5 Flash Lite Preview 09.2025",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 134.2,
   "n_obs": 19,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 276
  },
  {
   "key": "gemini 20 flash thinking experimental (jan '25)",
   "name": "Gemini 2.0 Flash Thinking Experimental (Jan '25)",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 134.2,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral large 2512",
   "name": "Mistral Large 2512",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 134.1,
   "n_obs": 14,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 277
  },
  {
   "key": "step3 vl 10b",
   "name": "Step3-VL-10B",
   "organization": "StepFun",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 134.1,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-01-15",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 20 pro exp 0205",
   "name": "Gemini 2.0 Pro Exp 02.05",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 134.0,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "olmo 332b think",
   "name": "Olmo 3 32B Think",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 133.9,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "granite 423b",
   "name": "Granite 4.2 3B",
   "organization": "IBM",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 133.9,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-25",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 278
  },
  {
   "key": "magistral medium 1",
   "name": "Magistral Medium 1",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 133.8,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "devstral 2",
   "name": "Devstral 2",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 133.7,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 279
  },
  {
   "key": "mistral medium 31",
   "name": "Mistral Medium 3.1",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 133.3,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 280
  },
  {
   "key": "grok 3 beta",
   "name": "Grok 3 Beta",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 133.2,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 281
  },
  {
   "key": "mistral large 3",
   "name": "Mistral Large 3",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 133.1,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-01",
   "cost_usd": 38.673735,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 282
  },
  {
   "key": "deepseek r10528 qwen38b",
   "name": "DeepSeek R1 0528 Qwen3 8B",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 132.9,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 20 flash thinking exp 0121",
   "name": "Gemini 2.0 Flash Thinking Exp 01.21",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 132.8,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral large 2",
   "name": "Mistral Large 2",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 132.8,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-07-24",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen38b",
   "name": "Qwen3 8B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 132.7,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-04-29",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 283
  },
  {
   "key": "ernie 45300b a47b",
   "name": "ERNIE 4.5 300B A47B",
   "organization": "Baidu",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 132.5,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "nanbeige413b",
   "name": "Nanbeige4.1-3B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 132.3,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "olmo 37b think",
   "name": "Olmo 3 7B Think",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 132.3,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "nemotron 3 nano omni 30b a3b reasoning",
   "name": "Nemotron 3 Nano Omni 30B A3B Reasoning",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 132.3,
   "n_obs": 8,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-04-28",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 284
  },
  {
   "key": "granite 4130b",
   "name": "Granite 4.1 30B",
   "organization": "IBM",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 132.2,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 285
  },
  {
   "key": "llama 31 nemotron nano 4b v11",
   "name": "Llama 3.1 Nemotron Nano 4B v1.1",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 132.1,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-05-20",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 20 pro experimental (feb '25)",
   "name": "Gemini 2.0 Pro Experimental (Feb '25)",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 132.1,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "minicpm51b",
   "name": "MiniCPM5-1B",
   "organization": "OpenBMB",
   "country": "China",
   "reasoning_status": "hybrid",
   "sci": 131.9,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-05-19",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 4 e4b",
   "name": "Gemma 4 E4B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 131.9,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 286
  },
  {
   "key": "devstral small 2",
   "name": "Devstral Small 2",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 131.8,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 287
  },
  {
   "key": "qwen34b 2507 instruct",
   "name": "Qwen3 4B 2507 Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 131.8,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 4 maverick 17b 128e instruct fp8",
   "name": "Llama-4-Maverick-17B-128E-Instruct-FP8",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 131.8,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral medium 2505",
   "name": "Mistral Medium 2505",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 131.7,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral medium 3",
   "name": "Mistral Medium 3",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 131.6,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "granite 418b",
   "name": "Granite 4.1 8B",
   "organization": "IBM",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 131.6,
   "n_obs": 8,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 288
  },
  {
   "key": "minicpm v 4613b",
   "name": "MiniCPM-V 4.6 1.3B",
   "organization": "OpenBMB",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 131.5,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek v3 (dec '24)",
   "name": "DeepSeek V3 (Dec '24)",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 131.5,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 289
  },
  {
   "key": "llama 4 maverick",
   "name": "Llama 4 Maverick",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 131.4,
   "n_obs": 8,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-04-05",
   "cost_usd": 62.28189309999999,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 290
  },
  {
   "key": "nvidia nemotron 3 nano 4b",
   "name": "NVIDIA Nemotron 3 Nano 4B",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 131.3,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-03-16",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 291
  },
  {
   "key": "claude 35 sonnet",
   "name": "Claude 3.5 Sonnet",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 131.3,
   "n_obs": 24,
   "n_configs": 2,
   "configs": [
    {
     "name": "Claude 3.5 Sonnet (20241022 snapshot)",
     "sci": 131.3,
     "n": 11,
     "headline": true
    },
    {
     "name": "Claude 3.5 Sonnet (standard)",
     "sci": 130.0,
     "n": 13
    }
   ],
   "release_date": "2024-06-20",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 292
  },
  {
   "key": "qwq 32b preview",
   "name": "QwQ-32B-Preview",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "reasoning",
   "sci": 131.2,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-11-28",
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek v3",
   "name": "DeepSeek-V3",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 131.2,
   "n_obs": 5,
   "n_configs": 2,
   "configs": [
    {
     "name": "DeepSeek-V3 (standard)",
     "sci": 131.2,
     "n": 5,
     "headline": true
    }
   ],
   "release_date": "2024-12-25",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 293
  },
  {
   "key": "gpt 4o (march 2025, chatgpt 4o latest)",
   "name": "GPT-4o (March 2025, chatgpt-4o-latest)",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 131.2,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 omni 30b a3b instruct",
   "name": "Qwen3 Omni 30B A3B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 131.2,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 20 flash",
   "name": "Gemini 2.0 Flash",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 131.1,
   "n_obs": 11,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-12-01",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 294
  },
  {
   "key": "sonar",
   "name": "Sonar",
   "organization": "Perplexity",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 130.9,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 20 flash (feb '25)",
   "name": "Gemini 2.0 Flash (Feb '25)",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 130.8,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 coder 30b a3b instruct",
   "name": "Qwen3 Coder 30B A3B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 130.7,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen25 max",
   "name": "Qwen2.5 Max",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 130.6,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 vl 8b",
   "name": "Qwen3 VL 8B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 130.6,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 20 flash (experimental)",
   "name": "Gemini 2.0 Flash (experimental)",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 130.5,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "g9v33b",
   "name": "G9v3-3B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 130.3,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "ministral 314b",
   "name": "Ministral 3 14B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 130.2,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 295
  },
  {
   "key": "reka flash 3",
   "name": "Reka Flash 3",
   "organization": "Reka",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 130.1,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 296
  },
  {
   "key": "ling mini 20",
   "name": "Ling-mini-2.0",
   "organization": "Ant Group",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 130.1,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "exaone 4012b",
   "name": "Exaone 4.0 1.2B",
   "organization": "LG AI Research",
   "country": "South Korea",
   "reasoning_status": "non_reasoning",
   "sci": 129.8,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 vl 8b instruct",
   "name": "Qwen3 VL 8B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 129.8,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-22",
   "cost_usd": 72.73489158,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 297
  },
  {
   "key": "deepseek v25",
   "name": "DeepSeek-V2.5",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 129.8,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-05-08",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 15 pro (sep '24)",
   "name": "Gemini 1.5 Pro (Sep '24)",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 129.7,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 4o (chatgpt)",
   "name": "GPT-4o (ChatGPT)",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 129.6,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral small 2506",
   "name": "Mistral Small 2506",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 129.6,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-06-20",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "grok 21212",
   "name": "Grok 2.1212",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 129.5,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-08-13",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 298
  },
  {
   "key": "grok 2 mini",
   "name": "Grok-2 mini",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 129.5,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-08-13",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 33 nemotron super 49b v1",
   "name": "Llama-3.3 Nemotron Super 49B v1",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 129.5,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-03-18",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 299
  },
  {
   "key": "mistral small 32",
   "name": "Mistral Small 3.2",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 129.4,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 300
  },
  {
   "key": "sonar pro",
   "name": "Sonar Pro",
   "organization": "Perplexity",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 129.3,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "magistral medium",
   "name": "Magistral Medium",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 129.2,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-06-10",
   "cost_usd": 40.236230400000004,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 301
  },
  {
   "key": "claude 35 sonnet (oct '24)",
   "name": "Claude 3.5 Sonnet (Oct '24)",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 129.2,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-10-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 327b",
   "name": "Gemma 3 27B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 129.2,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-03-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 31405b",
   "name": "Llama-3.1-405B",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 129.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "nova premier",
   "name": "Nova Premier",
   "organization": "Amazon",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 129.1,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 4o (aug '24)",
   "name": "GPT-4o (Aug '24)",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 129.1,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 302
  },
  {
   "key": "nova pro",
   "name": "Nova Pro",
   "organization": "Amazon",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 129.0,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-11-20",
   "cost_usd": 200.984304,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 303
  },
  {
   "key": "gemini 15 pro 001 feb24",
   "name": "Gemini 1.5 Pro 001 Feb24",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 129.0,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 4 maverick 17b 128e instruct",
   "name": "Llama-4-Maverick-17B-128E-Instruct",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 128.9,
   "n_obs": 11,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 304
  },
  {
   "key": "mistral large 3 (675b instruct 2512)",
   "name": "Mistral Large 3 (675B Instruct 2512)",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 128.8,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-04",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral large 3 (675b base)",
   "name": "Mistral Large 3 (675B Base)",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 128.8,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-04",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3 vl 4b instruct",
   "name": "Qwen3 VL 4B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 128.8,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-09-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 305
  },
  {
   "key": "llama 4 scout",
   "name": "Llama 4 Scout",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 128.8,
   "n_obs": 8,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-04-05",
   "cost_usd": 26.422708470000003,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 306
  },
  {
   "key": "qvq 72b preview",
   "name": "QvQ-72B-Preview",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 128.7,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-12-25",
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen max",
   "name": "Qwen Max",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 128.6,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-25",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "ministral 38b",
   "name": "Ministral 3 8B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 128.6,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 307
  },
  {
   "key": "gemma 3n e4b",
   "name": "Gemma 3n E4B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 128.5,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-06-26",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 308
  },
  {
   "key": "kimi k2 base",
   "name": "Kimi K2 Base",
   "organization": "Moonshot AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 128.5,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-07-11",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 4o (may '24)",
   "name": "GPT-4o (May '24)",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 128.5,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 309
  },
  {
   "key": "qwen 14b",
   "name": "Qwen-14B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 128.4,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": 18.79108813,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 20 flash lite (feb '25)",
   "name": "Gemini 2.0 Flash-Lite (Feb '25)",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 128.4,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 20 flash lite",
   "name": "Gemini 2.0 Flash-Lite",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 128.4,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-02-05",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen317b",
   "name": "Qwen3 1.7B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 128.3,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-04-29",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "ministral 3 (14b base 2512)",
   "name": "Ministral 3 (14B Base 2512)",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 128.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-04",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen2572b instruct",
   "name": "Qwen2.5 72B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 128.2,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-09-19",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 310
  },
  {
   "key": "lfm2526b",
   "name": "LFM2.5-2.6B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 128.1,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-04",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 311
  },
  {
   "key": "deepseek r1 distill llama 8b",
   "name": "DeepSeek R1 Distill Llama 8B",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 128.0,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-20",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "command a plus (052026)",
   "name": "Command A Plus (05-2026)",
   "organization": "Cohere",
   "country": "CA",
   "reasoning_status": "non_reasoning",
   "sci": 127.9,
   "n_obs": 15,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 312
  },
  {
   "key": "qwen3 vl 4b",
   "name": "Qwen3 VL 4B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 127.9,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "lfm258b a1b",
   "name": "LFM2.5-8B-A1B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 127.9,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 4o (nov '24)",
   "name": "GPT-4o (Nov '24)",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 127.8,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 15 pro",
   "name": "Gemini 1.5 Pro",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 127.7,
   "n_obs": 8,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-02-15",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 313
  },
  {
   "key": "nova lite",
   "name": "Nova Lite",
   "organization": "Amazon",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 127.6,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-11-20",
   "cost_usd": 21.164724179999997,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 314
  },
  {
   "key": "gpt 4o",
   "name": "GPT-4o",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 127.5,
   "n_obs": 21,
   "n_configs": 2,
   "configs": [
    {
     "name": "GPT-4o (standard)",
     "sci": 128.9,
     "n": 6,
     "thin": true
    },
    {
     "name": "GPT-4o (2024-11-20 snapshot)",
     "sci": 127.5,
     "n": 15,
     "headline": true
    }
   ],
   "release_date": "2024-05-13",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 315
  },
  {
   "key": "qwen25 vl 32b instruct",
   "name": "Qwen2.5 VL 32B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 127.4,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-02-28",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen25 vl 72b instruct",
   "name": "Qwen2.5-VL-72B-Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 127.4,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 4 turbo",
   "name": "GPT-4 Turbo",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 127.3,
   "n_obs": 5,
   "n_configs": 2,
   "configs": [
    {
     "name": "GPT-4 Turbo (2024-04-09 snapshot)",
     "sci": 127.3,
     "n": 5,
     "headline": true
    }
   ],
   "release_date": "2024-04-09",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 316
  },
  {
   "key": "kimi linear 48b a3b instruct",
   "name": "Kimi Linear 48B A3B Instruct",
   "organization": "Moonshot AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 127.3,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "command a",
   "name": "Command A",
   "organization": "Cohere",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 127.2,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-05-20",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 317
  },
  {
   "key": "llama 31405b instruct",
   "name": "Llama-3.1-405B-Instruct",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 127.2,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-07-23",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 318
  },
  {
   "key": "llama 3290b instruct",
   "name": "Llama 3.2 90B Instruct",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 127.2,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-09-25",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "olmo 3132b instruct",
   "name": "Olmo 3.1 32B Instruct",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 127.1,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral small 3224b instruct",
   "name": "Mistral Small 3.2 24B Instruct",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 127.1,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-06-20",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "lfm224b a2b",
   "name": "LFM2 24B A2B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 127.1,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 319
  },
  {
   "key": "llama 31 tulu3405b",
   "name": "Llama 3.1 Tulu3 405B",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 127.0,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "claude 35 sonnet (june '24)",
   "name": "Claude 3.5 Sonnet (June '24)",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 127.0,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 4 e2b",
   "name": "Gemma 4 E2B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "reasoning",
   "sci": 126.9,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 320
  },
  {
   "key": "gemini 15 flash (sep '24)",
   "name": "Gemini 1.5 Flash (Sep '24)",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 126.8,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral small 3124b instruct",
   "name": "Mistral Small 3.1 24B Instruct",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 126.8,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-03-17",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "phi 4",
   "name": "Phi 4",
   "organization": "Microsoft",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 126.7,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-12-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 321
  },
  {
   "key": "qwen2 vl 72b instruct",
   "name": "Qwen2-VL-72B-Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 126.7,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-08-29",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "grok 2 (dec '24)",
   "name": "Grok 2 (Dec '24)",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 126.6,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 327b it",
   "name": "Gemma 3 27B IT",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 126.4,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 322
  },
  {
   "key": "llama 3370b instruct",
   "name": "Llama-3.3-70B-Instruct",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 126.2,
   "n_obs": 10,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-12-06",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 323
  },
  {
   "key": "devstral small (may '25)",
   "name": "Devstral Small (May '25)",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 126.1,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 4o mini",
   "name": "GPT-4o Mini",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 126.1,
   "n_obs": 21,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-07-18",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 324
  },
  {
   "key": "qwen352b",
   "name": "Qwen3.5 2B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 126.1,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-03-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 325
  },
  {
   "key": "devstral medium",
   "name": "Devstral Medium",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 126.0,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-07-10",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral large 2 (nov '24)",
   "name": "Mistral Large 2 (Nov '24)",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 126.0,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen34b",
   "name": "Qwen3 4B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 126.0,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-04-29",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "claude 3 opus",
   "name": "Claude 3 Opus",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 126.0,
   "n_obs": 12,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-02-29",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 326
  },
  {
   "key": "ministral 3 (8b base 2512)",
   "name": "Ministral 3 (8B Base 2512)",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 125.9,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-04",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 327b instruct",
   "name": "Gemma 3 27B Instruct",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 125.8,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 327
  },
  {
   "key": "qwen2532b instruct",
   "name": "Qwen2.5 32B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 125.8,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-09-19",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "pixtral large",
   "name": "Pixtral Large",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 125.7,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-11-18",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 328
  },
  {
   "key": "grok beta",
   "name": "Grok Beta",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 125.7,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral large 2411",
   "name": "Mistral Large 2411",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 125.7,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen2514b instruct",
   "name": "Qwen2.5 14B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 125.5,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-09-19",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "olmo 37b instruct",
   "name": "Olmo 3 7B Instruct",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 125.3,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "sarvam m",
   "name": "Sarvam M",
   "organization": "Sarvam AI",
   "country": "India",
   "reasoning_status": "reasoning",
   "sci": 125.1,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-05-23",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 312b",
   "name": "Gemma 3 12B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 125.1,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-03-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 329
  },
  {
   "key": "gpt 41 nano",
   "name": "GPT-4.1 nano",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 125.0,
   "n_obs": 16,
   "n_configs": 2,
   "configs": [
    {
     "name": "GPT-4.1 nano (standard)",
     "sci": 126.2,
     "n": 5,
     "thin": true
    },
    {
     "name": "GPT-4.1 nano (2025-04-14 snapshot)",
     "sci": 125.0,
     "n": 11,
     "headline": true
    }
   ],
   "release_date": "2025-04-14",
   "cost_usd": 31.0257939,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 330
  },
  {
   "key": "devstral small (jul '25)",
   "name": "Devstral Small (Jul '25)",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 124.9,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 40125 preview",
   "name": "GPT-4.0125 Preview",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 124.8,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek coder v2 instruct",
   "name": "DeepSeek-Coder-V2-Instruct",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 124.7,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral small 31",
   "name": "Mistral Small 3.1",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 124.6,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 331
  },
  {
   "key": "gemini diffusion",
   "name": "Gemini Diffusion",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 124.5,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-05-20",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 41106 preview",
   "name": "GPT-4.1106 Preview",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 124.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen25 coder 32b instruct",
   "name": "Qwen2.5-Coder 32B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 124.4,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-09-19",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "ministral 33b",
   "name": "Ministral 3 3B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 124.4,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 332
  },
  {
   "key": "granite 40 h small",
   "name": "Granite 4.0 H Small",
   "organization": "IBM",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 124.4,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral large 2 (jul '24)",
   "name": "Mistral Large 2 (Jul '24)",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 124.3,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "lfm25 vl 3b",
   "name": "LFM2.5-VL-3B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 124.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 31 nemotron 70b instruct",
   "name": "Llama 3.1 Nemotron 70B Instruct",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 124.2,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-10-01",
   "cost_usd": 73.535496,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 312b instruct",
   "name": "Gemma 3 12B Instruct",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 124.2,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 333
  },
  {
   "key": "mistral nemo instruct",
   "name": "Mistral NeMo Instruct",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 124.1,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-07-18",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 4 scout 17b 16e instruct",
   "name": "Llama-4-Scout-17B-16E-Instruct",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 124.0,
   "n_obs": 12,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": 334
  },
  {
   "key": "gemini 15 flash 001",
   "name": "Gemini 1.5 Flash 001",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 124.0,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 3170b instruct",
   "name": "Llama-3.1-70B-Instruct",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 123.9,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-07-23",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 335
  },
  {
   "key": "pixtral 12b",
   "name": "Pixtral-12B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 123.6,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-09-17",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "yi lightning",
   "name": "Yi Lightning",
   "organization": "01.AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 123.6,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral large 2407",
   "name": "Mistral Large 2407",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 123.5,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral small 3",
   "name": "Mistral Small 3",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 123.4,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "claude 35 haiku",
   "name": "Claude 3.5 Haiku",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 123.3,
   "n_obs": 7,
   "n_configs": 2,
   "configs": [
    {
     "name": "Claude 3.5 Haiku (20241022 snapshot)",
     "sci": 123.3,
     "n": 7,
     "headline": true
    }
   ],
   "release_date": "2024-10-22",
   "cost_usd": 56.516124,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 336
  },
  {
   "key": "qwen25 turbo",
   "name": "Qwen2.5 Turbo",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 123.3,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral saba",
   "name": "Mistral Saba",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 123.3,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral small 2501",
   "name": "Mistral Small 2501",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 123.2,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-30",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "ministral 3 (3b base 2512)",
   "name": "Ministral 3 (3B Base 2512)",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 123.0,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-12-04",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral small 2503",
   "name": "Mistral Small 2503",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 122.9,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-03-17",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 40314",
   "name": "GPT-4.0314",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 122.8,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 15 pro (may '24)",
   "name": "Gemini 1.5 Pro (May '24)",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 122.7,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "phi 35 moe instruct",
   "name": "Phi-3.5-MoE-instruct",
   "organization": "Microsoft",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 122.7,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-08-23",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "yi large",
   "name": "Yi-large",
   "organization": "01.AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 122.6,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-05-13",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen25 vl 7b instruct",
   "name": "Qwen2.5 VL 7B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 122.6,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-26",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "jamba reasoning 3b",
   "name": "Jamba Reasoning 3B",
   "organization": "AI21",
   "country": "Israel",
   "reasoning_status": "reasoning",
   "sci": 122.3,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral small 3124b base",
   "name": "Mistral Small 3.1 24B Base",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 122.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-03-17",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen272b instruct",
   "name": "Qwen2 72B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 122.2,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-07-23",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "nova micro",
   "name": "Nova Micro",
   "organization": "Amazon",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 122.1,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-11-20",
   "cost_usd": 6.384766955,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 15 flash 8b",
   "name": "Gemini 1.5 Flash 8B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 122.1,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-03-15",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 3n e2b",
   "name": "Gemma 3n E2B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 122.0,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-06-26",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 337
  },
  {
   "key": "qwen25 omni 7b",
   "name": "Qwen2.5-Omni-7B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 121.9,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-03-27",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral small 324b base",
   "name": "Mistral Small 3 24B Base",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 121.9,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-30",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 15 flash",
   "name": "Gemini 1.5 Flash",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 121.9,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-05-01",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 32 instruct 90b (vision)",
   "name": "Llama 3.2 Instruct 90B (Vision)",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 121.7,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 338
  },
  {
   "key": "qwen257b instruct",
   "name": "Qwen2.5 7B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 121.6,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-09-19",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "amazon nova pro v1:0",
   "name": "amazon.nova-pro-v1:0",
   "organization": "Amazon",
   "country": "US",
   "reasoning_status": "unknown",
   "sci": 121.6,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "grok 15",
   "name": "Grok-1.5",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 121.5,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-03-28",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "grok 15v",
   "name": "Grok-1.5V",
   "organization": "xAI",
   "country": "US",
   "reasoning_status": "hybrid",
   "sci": 121.5,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-04-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "lfm2512b instruct",
   "name": "LFM2.5-1.2B-Instruct",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 120.8,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "jamba 17 large",
   "name": "Jamba 1.7 Large",
   "organization": "AI21",
   "country": "Israel",
   "reasoning_status": "non_reasoning",
   "sci": 120.7,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "amazon nova lite v1:0",
   "name": "amazon.nova-lite-v1:0",
   "organization": "Amazon",
   "country": "US",
   "reasoning_status": "unknown",
   "sci": 120.6,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 4",
   "name": "GPT-4",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 120.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2023-03-14",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "lfm2512b thinking",
   "name": "LFM2.5-1.2B-Thinking",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "reasoning",
   "sci": 120.4,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 3211b instruct",
   "name": "Llama 3.2 11B Instruct",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 120.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-09-25",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek vl2",
   "name": "DeepSeek-VL2",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 120.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-12-13",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek vl2 small",
   "name": "DeepSeek VL2 Small",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 120.0,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-12-13",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deephermes 3 mistral 24b preview",
   "name": "DeepHermes 3 - Mistral 24B Preview",
   "organization": "Nous Research",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 119.9,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 227b it",
   "name": "Gemma 2 27B IT",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 119.8,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "granite 413b",
   "name": "Granite 4.1 3B",
   "organization": "IBM",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 119.7,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "phi 35 vision instruct",
   "name": "Phi-3.5-vision-instruct",
   "organization": "Microsoft",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 119.6,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-08-23",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "phi 4 multimodal instruct",
   "name": "Phi-4-multimodal-instruct",
   "organization": "Microsoft",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 119.5,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-02-01",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "meta llama 370b instruct",
   "name": "Meta-Llama-3-70B-Instruct",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 119.3,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 3n e4b instruct preview (may '25)",
   "name": "Gemma 3n E4B Instruct Preview (May '25)",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 119.3,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "granite 338b base",
   "name": "Granite 3.3 8B Base",
   "organization": "IBM",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 119.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-04-16",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "hermes 370b",
   "name": "Hermes 3 70B",
   "organization": "Nous Research",
   "country": "US",
   "reasoning_status": "unknown",
   "sci": 119.2,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-08-15",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "claude 3 sonnet",
   "name": "Claude 3 Sonnet",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 119.1,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-02-29",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 339
  },
  {
   "key": "jamba 15 large",
   "name": "Jamba 1.5 Large",
   "organization": "AI21",
   "country": "Israel",
   "reasoning_status": "non_reasoning",
   "sci": 119.1,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-08-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "lfm28b a1b",
   "name": "LFM2 8B A1B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 119.1,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "jamba 16 large",
   "name": "Jamba 1.6 Large",
   "organization": "AI21",
   "country": "Israel",
   "reasoning_status": "non_reasoning",
   "sci": 119.0,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mixtral 8x22b instruct v01",
   "name": "Mixtral-8x22B-Instruct-v0.1",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 119.0,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 34b instruct",
   "name": "Gemma 3 4B Instruct",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 118.9,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 340
  },
  {
   "key": "hermes 3 llama 3170b",
   "name": "Hermes 3 - Llama-3.1 70B",
   "organization": "Nous Research",
   "country": "US",
   "reasoning_status": "unknown",
   "sci": 118.7,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-08-15",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 3n e4b instructed",
   "name": "Gemma 3n E4B Instructed",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 118.6,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-06-26",
   "cost_usd": 5.721499560000001,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 341
  },
  {
   "key": "phi 4 mini instruct",
   "name": "Phi-4 Mini Instruct",
   "organization": "Microsoft",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 118.5,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 342
  },
  {
   "key": "molmo28b",
   "name": "Molmo2-8B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 118.5,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek r1 distill qwen 15b",
   "name": "DeepSeek R1 Distill Qwen 1.5B",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "reasoning",
   "sci": 118.4,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-01-20",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 15 flash (may '24)",
   "name": "Gemini 1.5 Flash (May '24)",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 118.4,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek vl2 tiny",
   "name": "DeepSeek VL2 Tiny",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 118.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-12-13",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral small (feb '24)",
   "name": "Mistral Small (Feb '24)",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 118.1,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "inflection 1",
   "name": "Inflection-1",
   "organization": "Inflection",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 118.1,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "granite 338b",
   "name": "Granite 3.3 8B",
   "organization": "IBM",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 118.0,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral small (sep '24)",
   "name": "Mistral Small (Sep '24)",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 117.8,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen306b",
   "name": "Qwen3 0.6B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 117.8,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 40613",
   "name": "GPT-4.0613",
   "organization": "OpenAI",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 117.6,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 3 instruct 70b",
   "name": "Llama 3 Instruct 70B",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 117.5,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "c4ai command r plus 082024",
   "name": "C4ai Command R Plus 08.2024",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 117.5,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "palm 2 l",
   "name": "PaLM 2-L",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 117.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 29b it",
   "name": "Gemma 2 9B IT",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 117.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "phi 4 mini",
   "name": "Phi 4 Mini",
   "organization": "Microsoft",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 117.1,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-02-01",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen25 coder 7b instruct",
   "name": "Qwen2.5-Coder 7B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 117.1,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-09-19",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "granite 40 micro",
   "name": "Granite 4.0 Micro",
   "organization": "IBM",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 117.1,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 343
  },
  {
   "key": "hermes 3405b",
   "name": "Hermes 3 405B",
   "organization": "Nous Research",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 117.1,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-08-15",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "yi 1534b chat",
   "name": "Yi-1.5-34B-Chat",
   "organization": "01.AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 117.0,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen25 coder 14b",
   "name": "Qwen2.5-Coder-14B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 116.6,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "lfm226b",
   "name": "LFM2 2.6B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 116.6,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 34b",
   "name": "Gemma 3 4B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 116.4,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-03-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral large (feb '24)",
   "name": "Mistral Large (Feb '24)",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 116.4,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen1572b chat",
   "name": "Qwen1.5 72B Chat",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 116.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "olmo 232b",
   "name": "OLMo 2 32B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 116.0,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 227b",
   "name": "Gemma 2 27B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 115.9,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-06-27",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "phi 3 medium 128k instruct",
   "name": "Phi-3-medium-128k-instruct",
   "organization": "Microsoft",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 115.5,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "claude instant 12",
   "name": "Claude Instant 1.2",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 115.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "ministral 8b instruct",
   "name": "Ministral 8B Instruct",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 115.2,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-10-16",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "k2 horizon 09b",
   "name": "K2 Horizon 0.9B",
   "organization": "MBZUAI",
   "country": "UAE",
   "reasoning_status": "hybrid",
   "sci": 115.1,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-09-03",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 344
  },
  {
   "key": "phi 3 small 8k instruct",
   "name": "Phi-3-small-8k-instruct",
   "organization": "Microsoft",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 115.0,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "phi 35 mini instruct",
   "name": "Phi-3.5-mini-instruct",
   "organization": "Microsoft",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 115.0,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-08-23",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 10 pro",
   "name": "Gemini 1.0 Pro",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 114.9,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2023-12-06",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen27b instruct",
   "name": "Qwen2 7B Instruct",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 114.8,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-07-23",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 35 turbo",
   "name": "GPT-3.5 Turbo",
   "organization": "OpenAI",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 114.5,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2023-03-21",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 32 instruct 11b (vision)",
   "name": "Llama 3.2 Instruct 11B (Vision)",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 114.4,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 345
  },
  {
   "key": "llama 3 instruct 8b",
   "name": "Llama 3 Instruct 8B",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 114.4,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "stablebeluga2",
   "name": "StableBeluga2",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 114.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "jamba 15 mini",
   "name": "Jamba 1.5 Mini",
   "organization": "AI21",
   "country": "Israel",
   "reasoning_status": "non_reasoning",
   "sci": 114.1,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-08-22",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "claude 3 haiku",
   "name": "Claude 3 Haiku",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 113.9,
   "n_obs": 7,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-03-13",
   "cost_usd": 36.43772225,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 346
  },
  {
   "key": "lfm25 vl 16b",
   "name": "LFM2.5-VL-1.6B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 113.8,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 318b instruct",
   "name": "Llama-3.1-8B-Instruct",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 113.7,
   "n_obs": 9,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-07-23",
   "cost_usd": 27.707839700000005,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 347
  },
  {
   "key": "gemma 3n e2b instructed",
   "name": "Gemma 3n E2B Instructed",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 113.3,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-06-26",
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 348
  },
  {
   "key": "lfm 40b",
   "name": "LFM 40B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 113.3,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 323b instruct",
   "name": "Llama 3.2 3B Instruct",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 113.3,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-09-25",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "yi 159b chat",
   "name": "Yi-1.5-9B-Chat",
   "organization": "01.AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 113.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "claude 20",
   "name": "Claude 2.0",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 113.1,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "phi 3 mini instruct 38b",
   "name": "Phi-3 Mini Instruct 3.8B",
   "organization": "Microsoft",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 113.1,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 35 turbo 0125",
   "name": "GPT-3.5 Turbo 0125",
   "organization": "OpenAI",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 112.9,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 349
  },
  {
   "key": "gemma 29b",
   "name": "Gemma 2 9B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 112.8,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-06-27",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "palm 540b",
   "name": "PaLM 540B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 112.7,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "amazon nova micro v1:0",
   "name": "amazon.nova-micro-v1:0",
   "organization": "Amazon",
   "country": "US",
   "reasoning_status": "unknown",
   "sci": 112.6,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mixtral 8x7b v01",
   "name": "Mixtral-8x7B-v0.1",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 112.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "claude 21",
   "name": "Claude 2.1",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 112.4,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 3n e4b instructed litert preview",
   "name": "Gemma 3n E4B Instructed LiteRT Preview",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 112.2,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-05-20",
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "granite 401b",
   "name": "Granite 4.0 1B",
   "organization": "IBM",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 112.1,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "yi 34b",
   "name": "Yi-34B",
   "organization": "01.AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 111.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2023-11-02",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "falcon 180b",
   "name": "falcon-180B",
   "organization": "TII",
   "country": "UAE",
   "reasoning_status": "non_reasoning",
   "sci": 111.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2023-09-06",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "apertus 70b instruct",
   "name": "Apertus 70B Instruct",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 111.4,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "phi 3 mini 4k instruct",
   "name": "Phi-3-mini-4k-instruct",
   "organization": "Microsoft",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 111.3,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 35 turbo 1106",
   "name": "GPT-3.5 Turbo 1106",
   "organization": "OpenAI",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 111.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "claude instant 11",
   "name": "Claude Instant 1.1",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 111.1,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "ibm granite 40 tiny preview",
   "name": "IBM Granite 4.0 Tiny Preview",
   "organization": "IBM",
   "country": "US",
   "reasoning_status": "unknown",
   "sci": 111.0,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-05-02",
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "ministral 3b 2410",
   "name": "Ministral 3B 2410",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 110.8,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "wizardlm 28x22b",
   "name": "WizardLM-2-8x22B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 110.7,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek coder v2 lite instruct",
   "name": "DeepSeek-Coder-V2-Lite-Instruct",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 110.5,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemini 10 pro 001",
   "name": "Gemini 1.0 Pro 001",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 110.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "dbrx instruct",
   "name": "Dbrx Instruct",
   "organization": "Databricks",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 110.3,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "yi 34b chat",
   "name": "Yi-34B-Chat",
   "organization": "01.AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 109.8,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral nemo base 2407",
   "name": "Mistral-Nemo-Base-2407",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 109.8,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "chinchilla (70b)",
   "name": "Chinchilla (70B)",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 109.6,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "meta llama 38b instruct",
   "name": "Meta-Llama-3-8B-Instruct",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 109.3,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "text davinci 003",
   "name": "Text Davinci 003",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 109.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen25 coder 3b",
   "name": "Qwen2.5-Coder-3B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 108.7,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "command r+ (apr '24)",
   "name": "Command-R+ (Apr '24)",
   "organization": "Cohere",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 108.7,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "jamba 17 mini",
   "name": "Jamba 1.7 Mini",
   "organization": "AI21",
   "country": "Israel",
   "reasoning_status": "non_reasoning",
   "sci": 108.6,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "claude instant",
   "name": "Claude Instant",
   "organization": "Anthropic",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 108.2,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "palm 2 m",
   "name": "PaLM 2-M",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 108.1,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "north micro vision instruct",
   "name": "North Micro Vision Instruct",
   "organization": "Cohere",
   "country": "Canada",
   "reasoning_status": "non_reasoning",
   "sci": 108.0,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2026-08-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "c4ai command r 082024",
   "name": "C4ai Command R 08.2024",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 108.0,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mixtral 8x7b instruct v01",
   "name": "Mixtral-8x7B-Instruct-v0.1",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 107.8,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 31b instruct",
   "name": "Gemma 3 1B Instruct",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 107.8,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "apertus 8b instruct",
   "name": "Apertus 8B Instruct",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 107.7,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt 35 turbo 0613",
   "name": "GPT-3.5 Turbo 0613",
   "organization": "OpenAI",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 107.6,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "tiny aya global",
   "name": "Tiny Aya Global",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 107.5,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen1514b chat",
   "name": "Qwen1.5-14B-Chat",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 107.5,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 270b hf",
   "name": "Llama-2-70b-hf",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 107.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "text davinci 002",
   "name": "Text Davinci 002",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 107.0,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "openchat 35 (1210)",
   "name": "OpenChat 3.5 (1210)",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 106.4,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "yi 9b",
   "name": "Yi-9B",
   "organization": "01.AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 106.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "jamba 16 mini",
   "name": "Jamba 1.6 Mini",
   "organization": "AI21",
   "country": "Israel",
   "reasoning_status": "non_reasoning",
   "sci": 106.3,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "internlm 20b",
   "name": "Internlm 20B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 106.1,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "molmo 7b d",
   "name": "Molmo 7B-D",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 106.0,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 270b chat hf",
   "name": "Llama-2-70b-chat-hf",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 105.3,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "falcon 11b",
   "name": "Falcon 11B",
   "organization": "TII",
   "country": "UAE",
   "reasoning_status": "non_reasoning",
   "sci": 105.1,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deephermes 3 llama 318b preview",
   "name": "DeepHermes 3 - Llama-3.1 8B Preview",
   "organization": "Nous Research",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 104.8,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 3n e2b instructed litert (preview)",
   "name": "Gemma 3n E2B Instructed LiteRT (Preview)",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 104.7,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-05-20",
   "cost_usd": null,
   "prerelease": true,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "command r+",
   "name": "Command R+",
   "organization": "Cohere",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 104.6,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-08-30",
   "cost_usd": 299.181555,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 65b",
   "name": "LLaMA-65B",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 104.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2023-02-24",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "palm 2 s",
   "name": "PaLM 2-S",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 104.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek coder v2 lite base",
   "name": "DeepSeek-Coder-V2-Lite-Base",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 104.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen25 coder 15b",
   "name": "Qwen2.5-Coder-1.5B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 103.7,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "olmo 27b",
   "name": "OLMo 2 7B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 103.4,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen 7b",
   "name": "Qwen-7B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 103.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "nemotron 415b",
   "name": "Nemotron-4 15B",
   "organization": "NVIDIA",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 103.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral 7b instruct v03",
   "name": "Mistral-7B-Instruct-v0.3",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 103.1,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral 7b v01",
   "name": "Mistral-7B-v0.1",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 102.6,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 234b",
   "name": "Llama-2-34b",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 102.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "command r (mar '24)",
   "name": "Command-R (Mar '24)",
   "organization": "Cohere",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 102.1,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "glam (moe)",
   "name": "GLaM (MoE)",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 101.9,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "megatron turing nlg 530b",
   "name": "Megatron-Turing NLG 530B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 101.8,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "starcoder215b",
   "name": "Starcoder2 15B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 101.6,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 7b",
   "name": "Gemma 7B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 101.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "codeqwen157b",
   "name": "CodeQwen1.5-7B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 101.0,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen3508b",
   "name": "Qwen3.5 0.8B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 101.0,
   "n_obs": 6,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 350
  },
  {
   "key": "gopher (280b)",
   "name": "Gopher (280B)",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 100.9,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "lfm212b",
   "name": "LFM2 1.2B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 100.3,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek llm 67b chat",
   "name": "Deepseek Llm 67B Chat",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 100.2,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 213b chat",
   "name": "Llama-2-13b-chat",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 100.2,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "yi 6b chat",
   "name": "Yi-6B-Chat",
   "organization": "01.AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 100.1,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 33b",
   "name": "LLaMA-33B",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 100.1,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen157b chat",
   "name": "Qwen1.5-7B-Chat",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 99.7,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral 7b instruct v02",
   "name": "Mistral-7B-Instruct-v0.2",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 99.5,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "phi 2",
   "name": "Phi 2",
   "organization": "Microsoft",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 98.8,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "baichuan 213b base",
   "name": "Baichuan-2-13B-Base",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 98.7,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "text davinci 001",
   "name": "Text Davinci 001",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 98.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "palm 62b",
   "name": "PaLM 62B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 98.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek coder 33b base",
   "name": "Deepseek Coder 33B Base",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 98.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "command r plus",
   "name": "Command R Plus",
   "organization": "Cohere",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 98.1,
   "n_obs": 1,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "yi 6b",
   "name": "Yi-6B",
   "organization": "01.AI",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 98.0,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "baichuan213b chat",
   "name": "Baichuan2-13B-Chat",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 97.5,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mistral 7b instruct v01",
   "name": "Mistral-7B-Instruct-v0.1",
   "organization": "Mistral",
   "country": "France",
   "reasoning_status": "non_reasoning",
   "sci": 97.5,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "intellect 1 instruct",
   "name": "INTELLECT-1-Instruct",
   "organization": "Prime Intellect",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 97.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2024-11-29",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 31b",
   "name": "Gemma 3 1B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 97.0,
   "n_obs": 2,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-03-12",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 117b it",
   "name": "Gemma 1.1 7B IT",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 97.0,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "falcon 40b",
   "name": "Falcon 40B",
   "organization": "TII",
   "country": "UAE",
   "reasoning_status": "non_reasoning",
   "sci": 96.9,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2023-03-15",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek coder 67b base",
   "name": "Deepseek Coder 6.7B Base",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 96.9,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "internlm 7b",
   "name": "Internlm 7B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 96.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 321b instruct",
   "name": "Llama-3.2-1B-Instruct",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 95.9,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 351
  },
  {
   "key": "baichuan 13b base",
   "name": "Baichuan-13B-Base",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 95.8,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 13b",
   "name": "LLaMA-13B",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 95.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "internlm chat 20b",
   "name": "Internlm Chat 20B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 94.5,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "chatglm26b",
   "name": "Chatglm2 6B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 93.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "granite 40350m",
   "name": "Granite 4.0 350M",
   "organization": "IBM",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 93.1,
   "n_obs": 5,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": false,
   "provisional": false,
   "new_since": null,
   "badges": [
    "thin evidence"
   ],
   "rank": 352
  },
  {
   "key": "starcoder27b",
   "name": "Starcoder2 7B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 92.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "baichuan 27b base",
   "name": "Baichuan-2-7B-Base",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 91.8,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mpt 30b",
   "name": "Mpt 30B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 91.7,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "vicuna 13b v11",
   "name": "Vicuna 13B v1.1",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 91.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 27b",
   "name": "Llama-2-7b",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 89.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "starcoder23b",
   "name": "Starcoder2 3B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 88.9,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen25 coder 05b",
   "name": "Qwen2.5-Coder-0.5B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 88.5,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 7b it",
   "name": "Gemma 7B IT",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 88.5,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "deepseek coder 13b base",
   "name": "Deepseek Coder 1.3B Base",
   "organization": "DeepSeek",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 86.7,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "baichuan 7b",
   "name": "Baichuan-7B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 86.5,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "qwen 18b",
   "name": "Qwen-1_8B",
   "organization": "Alibaba",
   "country": "China",
   "reasoning_status": "unknown",
   "sci": 84.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 2 chat 7b",
   "name": "Llama 2 Chat 7B",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 83.3,
   "n_obs": 3,
   "n_configs": 1,
   "configs": [],
   "release_date": "2023-07-18",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "xgen 7b 8k base",
   "name": "Xgen 7B 8K Base",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 83.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "phi 15",
   "name": "Phi 1.5",
   "organization": "Microsoft",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 82.6,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 2b",
   "name": "Gemma 2B",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 82.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "llama 7b",
   "name": "LLaMA-7B",
   "organization": "Meta",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 81.1,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt neox 20b",
   "name": "GPT-Neox 20B",
   "organization": "OpenAI",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 80.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "falcon 7b",
   "name": "Falcon 7B",
   "organization": "TII",
   "country": "UAE",
   "reasoning_status": "non_reasoning",
   "sci": 79.7,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "open llama 7b",
   "name": "Open Llama 7B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 79.7,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 2b it",
   "name": "Gemma 2B IT",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 79.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "mpt 7b",
   "name": "Mpt 7B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 79.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 112b it",
   "name": "Gemma 1.1 2B IT",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 79.4,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "redpajama incite 7b base",
   "name": "RedPajama-INCITE-7B-Base",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "non_reasoning",
   "sci": 77.3,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gemma 3270m",
   "name": "Gemma 3 270M",
   "organization": "Google",
   "country": "US",
   "reasoning_status": "non_reasoning",
   "sci": 75.2,
   "n_obs": 4,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "opt 13b",
   "name": "Opt 13B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 74.3,
   "n_obs": 0,
   "n_configs": 2,
   "configs": [
    {
     "name": "Opt 13B (standard)",
     "sci": 74.3,
     "n": 0,
     "headline": true
    },
    {
     "name": "Opt 13B (standard)",
     "sci": 41.8,
     "n": 0
    }
   ],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "ernie 45",
   "name": "ERNIE 4.5",
   "organization": "Baidu",
   "country": "China",
   "reasoning_status": "non_reasoning",
   "sci": 74.1,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": "2025-06-25",
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt j 6b",
   "name": "GPT-J 6B",
   "organization": "OpenAI",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 73.1,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "dolly v212b",
   "name": "Dolly v2 12B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 68.7,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "cerebras gpt 13b",
   "name": "Cerebras-GPT-13B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 63.0,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt neo 27b",
   "name": "gpt-neo-2.7B",
   "organization": "OpenAI",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 51.0,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "gpt2 xl",
   "name": "Gpt2 Xl",
   "organization": "OpenAI",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 43.7,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  },
  {
   "key": "stablelm tuned alpha 7b",
   "name": "Stablelm Tuned Alpha 7B",
   "organization": "Unknown",
   "country": null,
   "reasoning_status": "unknown",
   "sci": 31.2,
   "n_obs": 0,
   "n_configs": 1,
   "configs": [],
   "release_date": null,
   "cost_usd": null,
   "prerelease": false,
   "hidden": true,
   "provisional": false,
   "new_since": null,
   "badges": [],
   "rank": null
  }
 ],
 "inflation": {
  "labs": [
   {
    "org": "Amazon",
    "median_gap": 12.5,
    "mean_gap": 10.32,
    "n": 3,
    "n_suspect": 6
   },
   {
    "org": "Cohere",
    "median_gap": 6.16,
    "mean_gap": 6.16,
    "n": 2,
    "n_suspect": 1
   },
   {
    "org": "Xiaomi",
    "median_gap": 6.07,
    "mean_gap": 5.72,
    "n": 5,
    "n_suspect": 4
   },
   {
    "org": "Poolside",
    "median_gap": 5.62,
    "mean_gap": 5.62,
    "n": 2,
    "n_suspect": 0
   },
   {
    "org": "Tencent",
    "median_gap": 5.49,
    "mean_gap": 6.01,
    "n": 4,
    "n_suspect": 0
   },
   {
    "org": "Moonshot AI",
    "median_gap": 2.83,
    "mean_gap": 3.68,
    "n": 41,
    "n_suspect": 6
   },
   {
    "org": "IBM",
    "median_gap": 2.78,
    "mean_gap": 2.78,
    "n": 2,
    "n_suspect": 0
   },
   {
    "org": "Zhipu AI",
    "median_gap": 2.65,
    "mean_gap": 3.27,
    "n": 47,
    "n_suspect": 8
   },
   {
    "org": "Mistral",
    "median_gap": 2.21,
    "mean_gap": 2.99,
    "n": 4,
    "n_suspect": 1
   },
   {
    "org": "NVIDIA",
    "median_gap": 2.01,
    "mean_gap": 3.35,
    "n": 28,
    "n_suspect": 2
   },
   {
    "org": "DeepSeek",
    "median_gap": 1.3,
    "mean_gap": 3.22,
    "n": 39,
    "n_suspect": 16
   },
   {
    "org": "AI21",
    "median_gap": 1.14,
    "mean_gap": 0.47,
    "n": 4,
    "n_suspect": 0
   },
   {
    "org": "Alibaba",
    "median_gap": 1.12,
    "mean_gap": 1.9,
    "n": 125,
    "n_suspect": 22
   },
   {
    "org": "OpenAI",
    "median_gap": 0.98,
    "mean_gap": 2.37,
    "n": 68,
    "n_suspect": 28
   },
   {
    "org": "xAI",
    "median_gap": 0.56,
    "mean_gap": 3.57,
    "n": 20,
    "n_suspect": 4
   },
   {
    "org": "Google",
    "median_gap": 0.12,
    "mean_gap": 1.4,
    "n": 86,
    "n_suspect": 7
   },
   {
    "org": "Meta",
    "median_gap": 0.1,
    "mean_gap": 1.21,
    "n": 30,
    "n_suspect": 7
   },
   {
    "org": "Thinking Machines",
    "median_gap": 0.01,
    "mean_gap": 2.29,
    "n": 4,
    "n_suspect": 0
   },
   {
    "org": "Anthropic",
    "median_gap": 0.0,
    "mean_gap": 1.67,
    "n": 78,
    "n_suspect": 9
   },
   {
    "org": "Microsoft",
    "median_gap": 0.0,
    "mean_gap": 0.52,
    "n": 7,
    "n_suspect": 1
   },
   {
    "org": "MiniMax",
    "median_gap": -0.1,
    "mean_gap": 0.48,
    "n": 31,
    "n_suspect": 1
   },
   {
    "org": "Inception",
    "median_gap": -0.2,
    "mean_gap": -0.67,
    "n": 3,
    "n_suspect": 0
   },
   {
    "org": "Upstage",
    "median_gap": -0.82,
    "mean_gap": -0.82,
    "n": 2,
    "n_suspect": 0
   },
   {
    "org": "StepFun",
    "median_gap": -1.03,
    "mean_gap": -1.03,
    "n": 1,
    "n_suspect": 0
   },
   {
    "org": "LG AI Research",
    "median_gap": -1.07,
    "mean_gap": -1.07,
    "n": 1,
    "n_suspect": 0
   },
   {
    "org": "MBZUAI",
    "median_gap": -2.83,
    "mean_gap": -2.81,
    "n": 4,
    "n_suspect": 0
   },
   {
    "org": "Ant Group",
    "median_gap": -7.9,
    "mean_gap": -7.9,
    "n": 1,
    "n_suspect": 0
   }
  ],
  "unknown": {
   "median_gap": 2.6,
   "n": 21,
   "n_suspect": 3
  },
  "span": 12.5,
  "pairs": [
   {
    "model": "MiniMax-M2.1",
    "organization": "MiniMax",
    "benchmark": "Terminal-Bench",
    "self": 47.9,
    "third": 32.9,
    "gap": 15.0,
    "tier": "silver"
   },
   {
    "model": "Llama-3.3 Nemotron Super 49B v1",
    "organization": "NVIDIA",
    "benchmark": "GPQA",
    "self": 66.7,
    "third": 51.7,
    "gap": 15.0,
    "tier": "silver"
   },
   {
    "model": "Grok 4",
    "organization": "xAI",
    "benchmark": "Humanity's Last Exam",
    "self": 40.0,
    "third": 25.3,
    "gap": 14.7,
    "tier": "silver"
   },
   {
    "model": "GPT-6 Astra",
    "organization": "OpenAI",
    "benchmark": "FrontierMath Tier 4 (v2)",
    "self": 97.6,
    "third": 82.9,
    "gap": 14.7,
    "tier": "silver"
   },
   {
    "model": "GPT-5.6 Sol",
    "organization": "OpenAI",
    "benchmark": "Terminal-Bench 2.1",
    "self": 88.8,
    "third": 74.2,
    "gap": 14.6,
    "tier": "silver"
   },
   {
    "model": "Qwen3.6 27B (Non-reasoning)",
    "organization": "Alibaba",
    "benchmark": "Terminal-Bench 2.0",
    "self": 59.3,
    "third": 44.9,
    "gap": 14.4,
    "tier": "silver"
   },
   {
    "model": "Nova Pro",
    "organization": "Amazon",
    "benchmark": "GPQA",
    "self": 64.1,
    "third": 49.9,
    "gap": 14.2,
    "tier": "silver"
   },
   {
    "model": "GPT-5.6 Luna",
    "organization": "OpenAI",
    "benchmark": "GPQA",
    "self": 92.3,
    "third": 78.1,
    "gap": 14.2,
    "tier": "silver"
   },
   {
    "model": "Gemini 3 Flash",
    "organization": "Google",
    "benchmark": "Humanity's Last Exam",
    "self": 38.6,
    "third": 24.8,
    "gap": 13.8,
    "tier": "silver"
   },
   {
    "model": "DeepSeek V3.2",
    "organization": "DeepSeek",
    "benchmark": "LiveCodeBench",
    "self": 83.3,
    "third": 69.9,
    "gap": 13.4,
    "tier": "silver"
   },
   {
    "model": "Claude 3.7 Sonnet",
    "organization": "Anthropic",
    "benchmark": "GPQA",
    "self": 84.8,
    "third": 71.4,
    "gap": 13.4,
    "tier": "silver"
   },
   {
    "model": "GLM-4.6",
    "organization": "Zhipu AI",
    "benchmark": "LiveCodeBench",
    "self": 82.8,
    "third": 69.5,
    "gap": 13.3,
    "tier": "silver"
   },
   {
    "model": "GPT-5.3 Codex",
    "organization": "OpenAI",
    "benchmark": "Terminal-Bench 2.0",
    "self": 77.3,
    "third": 64.0,
    "gap": 13.3,
    "tier": "silver"
   },
   {
    "model": "Nemotron 3 Nano (30B A3B)",
    "organization": "NVIDIA",
    "benchmark": "LiveCodeBench",
    "self": 68.3,
    "third": 55.0,
    "gap": 13.3,
    "tier": "silver"
   },
   {
    "model": "Hy3-preview (Non-reasoning)",
    "organization": "Tencent",
    "benchmark": "AA-LCR",
    "self": 73.4,
    "third": 60.3,
    "gap": 13.1,
    "tier": "silver"
   },
   {
    "model": "Grok 4 Fast",
    "organization": "xAI",
    "benchmark": "GPQA",
    "self": 85.7,
    "third": 72.7,
    "gap": 13.0,
    "tier": "silver"
   },
   {
    "model": "Qwen3.5 122B A10B (Non-reasoning)",
    "organization": "Alibaba",
    "benchmark": "IFBench",
    "self": 76.1,
    "third": 63.3,
    "gap": 12.8,
    "tier": "silver"
   },
   {
    "model": "Grok-3",
    "organization": "xAI",
    "benchmark": "GPQA",
    "self": 84.6,
    "third": 71.8,
    "gap": 12.8,
    "tier": "silver"
   },
   {
    "model": "Qwen3.7 Max",
    "organization": "Alibaba",
    "benchmark": "LiveCodeBench v6",
    "self": 91.6,
    "third": 78.9,
    "gap": 12.7,
    "tier": "silver"
   },
   {
    "model": "DeepSeek V3.2",
    "organization": "DeepSeek",
    "benchmark": "Humanity's Last Exam",
    "self": 25.1,
    "third": 12.5,
    "gap": 12.6,
    "tier": "silver"
   },
   {
    "model": "Qwen3.5 0.8B (Non-reasoning)",
    "organization": "Alibaba",
    "benchmark": "MMLU-Pro",
    "self": 42.3,
    "third": 29.7,
    "gap": 12.6,
    "tier": "silver"
   },
   {
    "model": "grok-2-1212",
    "organization": "xAI",
    "benchmark": "MATH",
    "self": 76.1,
    "third": 63.5,
    "gap": 12.6,
    "tier": "silver"
   },
   {
    "model": "GPT-6 Astra",
    "organization": "OpenAI",
    "benchmark": "ARC-AGI",
    "self": 98.5,
    "third": 86.0,
    "gap": 12.5,
    "tier": "silver"
   },
   {
    "model": "Nova Pro",
    "organization": "Amazon",
    "benchmark": "MMLU-Pro",
    "self": 81.6,
    "third": 69.1,
    "gap": 12.5,
    "tier": "silver"
   },
   {
    "model": "MiniMax-M3.1-Flash",
    "organization": "MiniMax",
    "benchmark": "Terminal-Bench 4.0",
    "self": 38.1,
    "third": 25.8,
    "gap": 12.3,
    "tier": "silver"
   },
   {
    "model": "GLM-5.1",
    "organization": "Zhipu AI",
    "benchmark": "Terminal-Bench 2.0",
    "self": 66.2,
    "third": 53.9,
    "gap": 12.3,
    "tier": "silver"
   },
   {
    "model": "Qwen3.5 9B (Non-reasoning)",
    "organization": "Alibaba",
    "benchmark": "IFBench",
    "self": 64.5,
    "third": 52.3,
    "gap": 12.2,
    "tier": "silver"
   },
   {
    "model": "Claude Opus 4.8",
    "organization": "Anthropic",
    "benchmark": "FrontierCode Main",
    "self": 46.5,
    "third": 34.3,
    "gap": 12.2,
    "tier": "silver"
   },
   {
    "model": "Command A",
    "organization": "Cohere",
    "benchmark": "Terminal-Bench (hard subset)",
    "self": 25.0,
    "third": 12.9,
    "gap": 12.1,
    "tier": "silver"
   },
   {
    "model": "Qwen3-Coder 480B A35B Instruct",
    "organization": "Alibaba",
    "benchmark": "Terminal-Bench",
    "self": 37.5,
    "third": 25.4,
    "gap": 12.1,
    "tier": "silver"
   }
  ],
  "tier_counts": {
   "gold": 450,
   "silver": 213,
   "bronze": 128
  }
 },
 "under_review": []
}