{
 "list": "AI Models & Benchmarks",
 "slug": "models",
 "url": "https://ai.justout.today/models/",
 "publisher": "JustOut Today",
 "tagline": "Where to compare models: leaderboards, the benchmarks behind them, and each model family's official page.",
 "categories": [
  "Leaderboards",
  "Reasoning & knowledge benchmarks",
  "Coding & agent benchmarks",
  "Model families (official pages)",
  "Research trackers"
 ],
 "built": "2026-11-20T00:17:41-08:00",
 "last_walkthrough": "2026-11-20T00:03:30-08:00",
 "entries": [
  {
   "id": "lmarena",
   "name": "LMArena",
   "category": "Leaderboards",
   "description": "Crowd-voted head-to-head model rankings.",
   "url": "https://arena.ai",
   "note": "lmarena.ai now redirects here",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "artificial-analysis",
   "name": "Artificial Analysis",
   "category": "Leaderboards",
   "description": "Independent comparisons of model quality, speed and price.",
   "url": "https://artificialanalysis.ai",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "livebench",
   "name": "LiveBench",
   "category": "Leaderboards",
   "description": "Benchmark with fresh questions to limit contamination.",
   "url": "https://livebench.ai",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "aider-llm-leaderboards",
   "name": "Aider LLM leaderboards",
   "category": "Leaderboards",
   "description": "How well models edit code in Aider's polyglot benchmark.",
   "url": "https://aider.chat/docs/leaderboards/",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "helm",
   "name": "HELM",
   "category": "Leaderboards",
   "description": "Stanford's holistic evaluation of language models.",
   "url": "https://crfm.stanford.edu/helm/",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "mteb-leaderboard",
   "name": "MTEB leaderboard",
   "category": "Leaderboards",
   "description": "Massive Text Embedding Benchmark rankings.",
   "url": "https://huggingface.co/spaces/mteb/leaderboard",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "mmlu",
   "name": "MMLU",
   "category": "Reasoning & knowledge benchmarks",
   "description": "57-subject multiple-choice knowledge test.",
   "url": "https://arxiv.org/abs/2009.03300",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "gpqa",
   "name": "GPQA",
   "category": "Reasoning & knowledge benchmarks",
   "description": "Graduate-level, 'Google-proof' science questions.",
   "url": "https://arxiv.org/abs/2311.12022",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "humanity-s-last-exam",
   "name": "Humanity's Last Exam",
   "category": "Reasoning & knowledge benchmarks",
   "description": "Very hard expert-written questions across many fields.",
   "url": "https://lastexam.ai",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "arc-agi",
   "name": "ARC-AGI",
   "category": "Reasoning & knowledge benchmarks",
   "description": "Abstract reasoning puzzles easy for people, hard for AI.",
   "url": "https://arcprize.org",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "frontiermath",
   "name": "FrontierMath",
   "category": "Reasoning & knowledge benchmarks",
   "description": "Unpublished research-level maths problems.",
   "url": "https://epoch.ai/frontiermath",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "math",
   "name": "MATH",
   "category": "Reasoning & knowledge benchmarks",
   "description": "Competition mathematics problems with step-by-step solutions.",
   "url": "https://arxiv.org/abs/2103.03874",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "gsm8k",
   "name": "GSM8K",
   "category": "Reasoning & knowledge benchmarks",
   "description": "Grade-school maths word problems.",
   "url": "https://github.com/openai/grade-school-math",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "mmmu",
   "name": "MMMU",
   "category": "Reasoning & knowledge benchmarks",
   "description": "College-level multimodal understanding and reasoning.",
   "url": "https://mmmu-benchmark.github.io",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "simpleqa",
   "name": "SimpleQA",
   "category": "Reasoning & knowledge benchmarks",
   "description": "Short factual questions to measure hallucination.",
   "url": "https://openai.com/index/introducing-simpleqa/",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "status": "bot-blocked"
  },
  {
   "id": "swe-bench",
   "name": "SWE-bench",
   "category": "Coding & agent benchmarks",
   "description": "Can a model resolve real GitHub issues?",
   "url": "https://www.swebench.com",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "humaneval",
   "name": "HumanEval",
   "category": "Coding & agent benchmarks",
   "description": "Classic Python function-writing benchmark.",
   "url": "https://github.com/openai/human-eval",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "terminal-bench",
   "name": "Terminal-Bench",
   "category": "Coding & agent benchmarks",
   "description": "Agents completing real tasks in a terminal.",
   "url": "https://www.tbench.ai",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "tau-bench",
   "name": "tau-bench",
   "category": "Coding & agent benchmarks",
   "description": "Tool-using agents in customer-service style tasks.",
   "url": "https://github.com/sierra-research/tau-bench",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "osworld",
   "name": "OSWorld",
   "category": "Coding & agent benchmarks",
   "description": "Agents operating real desktop computers.",
   "url": "https://arxiv.org/abs/2404.07972",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "webarena",
   "name": "WebArena",
   "category": "Coding & agent benchmarks",
   "description": "Agents completing tasks on realistic websites.",
   "url": "https://webarena.dev",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "gaia",
   "name": "GAIA",
   "category": "Coding & agent benchmarks",
   "description": "General AI assistant tasks needing tools and browsing.",
   "url": "https://arxiv.org/abs/2311.12983",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "browsecomp",
   "name": "BrowseComp",
   "category": "Coding & agent benchmarks",
   "description": "Hard-to-find facts that require persistent web browsing.",
   "url": "https://openai.com/index/browsecomp/",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "status": "bot-blocked"
  },
  {
   "id": "openai-models",
   "name": "OpenAI models",
   "category": "Model families (official pages)",
   "description": "Current OpenAI model list and capabilities.",
   "url": "https://developers.openai.com/api/docs/models",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "claude-models",
   "name": "Claude models",
   "category": "Model families (official pages)",
   "description": "Current Claude model list.",
   "url": "https://platform.claude.com/docs/en/models/overview",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "gemini-models",
   "name": "Gemini models",
   "category": "Model families (official pages)",
   "description": "Current Gemini API models.",
   "url": "https://aistudio.google.com/docs/models",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "gemma",
   "name": "Gemma",
   "category": "Model families (official pages)",
   "description": "Google's open-weight model family.",
   "url": "https://deepmind.google/models/gemma/",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "llama",
   "name": "Llama",
   "category": "Model families (official pages)",
   "description": "Meta's open-weight model family.",
   "url": "https://dev.meta.ai/",
   "note": "llama.com now redirects here",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "mistral-models",
   "name": "Mistral models",
   "category": "Model families (official pages)",
   "description": "Mistral's open and commercial models.",
   "url": "https://docs.mistral.ai/getting-started/models/",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "qwen",
   "name": "Qwen",
   "category": "Model families (official pages)",
   "description": "Alibaba's open model family on GitHub.",
   "url": "https://github.com/QwenLM",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "deepseek",
   "name": "DeepSeek",
   "category": "Model families (official pages)",
   "description": "DeepSeek's open model releases on GitHub.",
   "url": "https://github.com/deepseek-ai",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "epoch-ai",
   "name": "Epoch AI",
   "category": "Research trackers",
   "description": "Data on AI compute, models and trends.",
   "url": "https://epoch.ai",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  },
  {
   "id": "metr",
   "name": "METR",
   "category": "Research trackers",
   "description": "Measures how long a task AI agents can finish on their own.",
   "url": "https://metr.org",
   "first_seen": "2026-11-19T23:58:13-08:00",
   "last_verified": "2026-11-20T00:03:30-08:00",
   "verified_by": "verify.py",
   "status": "verified"
  }
 ]
}
