LMArena · GPQA · SWE-bench · HLE · ARC-AGI · OSWorld · BFCL
Бенчмарки ИИ-моделей
3086 результатов 1079 моделей на 23 открытых бенчмарках. Каждое число ведёт к источнику; данные от разработчиков отмечены отдельно.
3086 результатов 1079 моделей 23 бенчмарков Обновлено 22 окт. 2026 г.
Current champions
The top model on each core benchmark. Only official leaderboards, independent evaluations and compiled data count; vendor self-reports are excluded.
LMArena Text Gemini 4 Argon 1525 Предпочтения людей GPQA Diamond Gemini 4 Argon 96.4% Наука Humanity's Last Exam GPT-6 Astra 54.8% Знания AIME (OTIS Mock AIME 2024–2025) Gemini 4 Argon 100.0% Математика SWE-bench Verified Claude Opus 4.7 83.5% Агентное программирование Terminal-Bench 2.0 GPT-5.5 84.7% Агентное программирование OSWorld (Verified) Claude Sonnet 4.6 72.1% Работа с компьютером τ²-bench Gemini 3.0 Pro 85.3% Инструменты и агенты BFCL v4 (function calling) Claude Opus 4.5 77.5% Инструменты и агенты MMLU-Pro Gemini 3.1 Pro 91.2% Знания
Матрица «модель × бенчмарк»
В каждой ячейке — лучший результат в основной настройке (приоритет у официальных рейтингов и независимых тестов). ◇ — только данные разработчика или статьи. Нажмите на столбец для сортировки, отметьте модели для сравнения.
| Модель | LMArena Text | GPQA Diamond | Humanity's Last Exam | AIME (OTIS Mock AIME 2024–2025) | SWE-bench Verified | Terminal-Bench 2.0 | OSWorld (Verified) | τ²-bench | BFCL v4 (function calling) | MMLU-Pro |
|---|---|---|---|---|---|---|---|---|---|---|
| GPT-5.2OpenAI | 1437 | 91.4% | 27.8% | 96.1% | 73.8% | 64.9% | — | 81.6% | 55.9% | 87.4% ◇ |
| Gemini 3 ProGoogle | 1486 | 92.6% | 37.5% | 91.4% | 77.4% | 69.4% | — | 75.9% | 72.5% | 90.1% ◇ |
| Claude Sonnet 4.5Anthropic | 1456 | 82.3% | 13.7% | 77.8% | 74.8% | 46.5% | 62.9% | 72.4% | 73.2% | — |
| GPT-5OpenAI | 1435 | 86.2% | 25.3% | 91.4% | 74.4% | 49.6% | — | 81.6% | — | 87.1% ◇ |
| o3OpenAI | 1432 | 81.8% | 20.3% | 84.4% | 62.3% | — | 23.0% | 73.9% ◇ | 63.0% | — |
| Claude Opus 4.5Anthropic | 1474 | 86.0% | 25.2% | 86.1% | 79.2% | 63.1% | 66.3% ◇ | 79.6% | 77.5% | — |
| Gemini 2.5 Pro (Jun 2025)Google | 1446 | 85.3% | 21.6% | 84.7% | 57.6% | 32.6% | — | — | — | 86.0% ◇ |
| Claude Sonnet 4Anthropic | 1402 | 79.2% | 7.8% | 68.9% | 76.8% | — | 43.9% | 80.5% ◇ | — | — |
| o4-miniOpenAI | 1391 | 79.6% | 18.1% | 81.7% | 64.6% | — | — | 68.3% | 53.2% | — |
| GPT-4.1OpenAI | 1415 | 66.9% | 5.4% | 38.3% | 48.5% | — | — | 74.0% | 54.0% | 81.8% ◇ |
| Claude 3.7 SonnetAnthropic | 1388 | 79.7% | 8.0% | 57.8% | 66.4% | — | 35.8% | 72.1% | — | 84.0% ◇ |
| Qwen2.5-72BAlibaba (Qwen) | 1303 | 49.1% | — | 8.1% | — | — | 5.0% | — | — | 71.6% ◇ |
| Claude Opus 4.7Anthropic | 1501 | 90.2% | 36.2% | 97.8% | 83.5% | 80.2% | — | — | — | — |
| GPT-5.4OpenAI | 1475 | 93.3% | 36.2% | 97.8% | 76.9% | 81.8% | — | — | — | 87.5% ◇ |
| Gemini 3.1 ProGoogle | 1487 | 94.4% | 46.4% | 95.6% | 75.6% | 80.2% | — | — | — | 91.2% |
| Claude Sonnet 4.6Anthropic | 1472 | 87.4% | — | 85.8% | 75.2% | 53.4% | 72.1% | — | — | — |
| Kimi K2.5Moonshot AI | 1450 | 87.6% | 24.4% | 92.2% | 73.8% | 43.2% | 63.3% | — | — | — |
| GPT-5.1OpenAI | 1456 | 87.6% | 23.7% | 88.6% | 68.0% | 47.6% | — | — | — | 86.4% ◇ |
| GPT-5 miniOpenAI | 1390 | 75.0% | 19.4% | 86.7% | 64.7% | 34.8% | — | — | 55.5% | — |
| Claude Opus 4Anthropic | 1426 | 76.3% | 10.7% | 64.4% | 70.7% | — | — | 81.4% ◇ | — | — |
| Gemini 1.5 Pro (Sept 2024)Google | 1351 | 57.2% | 4.6% | 23.1% | — | — | — | — | — | 70.3% |
| Claude 3.5 SonnetAnthropic | 1374 | 54.0% | — | 6.5% | 63.4% | — | — | — | — | 77.6% |
| GPT-4o (May 2024)OpenAI | 1346 | 48.9% | — | 6.3% | 38.8% | — | — | — | — | 72.5% |
| Claude 3 HaikuAnthropic | 1262 | 36.3% | — | 1.8% | 40.6% | — | — | — | — | 42.3% |
| GPT-4 Turbo (Nov 2023)OpenAI | 1325 | 42.4% | — | — | 22.4% | — | — | — | — | 63.7% |
| GLM-5Zhipu AI (Z.ai) | 1458 | 87.8% | — | 80.0% | 72.8% | 52.4% | — | 73.7% | — | 86.0% ◇ |
| Claude Opus 4.6Anthropic | 1504 | 90.5% | 34.4% | 94.4% | 78.7% | 79.8% | — | — | — | — |
| Gemini 3 FlashGoogle | 1473 | 89.4% | — | 95.6% | 75.8% | 64.3% | — | 76.8% | — | — |
| DeepSeek-V3.2DeepSeek | 1425 | 83.4% | — | 87.8% | 70.0% | 39.5% | — | 81.1% ◇ | — | 85.0% ◇ |
| Claude Haiku 4.5Anthropic | 1414 | 71.2% | — | 66.7% | 66.6% | 35.5% | — | — | 68.7% | — |
| GPT-5 nanoOpenAI | 1337 | 69.4% | — | 81.1% | 34.8% | 21.8% | — | — | 51.5% | — |
| Claude Opus 4.1Anthropic | 1450 | 76.8% | 11.5% | 68.9% | 73.3% | 38.0% | — | 82.4% ◇ | — | — |
| GPT-4.1 miniOpenAI | 1383 | 65.8% | — | 44.7% | 23.9% | — | — | 61.4% | 50.5% | — |
| Phi-4Microsoft Research | 1256 | 56.1% | — | 13.8% | — | — | — | — | 28.8% | 70.4% ◇ |
| Llama 3.3 70BMeta | 1318 | 47.4% | — | 5.1% | — | — | — | — | 31.9% | 65.9% |
| Gemini 1.5 Flash (Sep 2024)Google | 1309 | 47.3% | — | 16.3% | — | — | — | — | — | 64.1% |
| Llama 3.1-8BMeta | 1211 | 27.0% | — | 1.7% | — | — | — | — | 25.8% | 36.6% |
| GPT-4o miniOpenAI | 1318 | 37.7% | — | 6.9% | — | — | — | — | — | 63.1% |
| Claude 3 OpusAnthropic | 1322 | 47.2% | — | 4.7% | 15.8% | — | — | — | — | 68.5% |
| Gemini 3.8 FlashGoogle | 1497 | 95.4% | 44.5% | 98.9% | — | — | — | — | — | — |
Все бенчмарки
Программирование
LMArena WebDevАктивен Arena where two models each build a web app from the same prompt and users vote on the result. 127 моделей · Лучший: 1813 LiveCodeBenchДавно не обновлялся Coding problems collected continuously from LeetCode, AtCoder and Codeforces with their release dates, so models can be scored only on problems published after their training cut-off. 22 моделей · Лучший: 75.8% HumanEval / HumanEval+Насыщен 164 hand-written Python function problems from OpenAI (2021). EvalPlus adds many more tests per problem (HumanEval+) to catch wrong solutions that pass the original tests. 122 моделей · Лучший: 89.0%
Мультимодальность
Общее
Open LLM Leaderboard v2В архиве Hugging Face's leaderboard of open-weight models, scored on IFEval, BBH, MATH Lvl 5, GPQA, MuSR and MMLU-Pro with normalised scores. The leaderboard was archived in 2025 and no longer accepts submissions. 294 моделей · Лучший: 48.0% Artificial Analysis Intelligence IndexТолько ссылка Independent composite index that runs a battery of evaluations plus price and speed measurements across API models. Только ссылка
Знания
MMLUНасыщен About 14,000 multiple-choice questions over 57 subjects, from elementary maths to law and medicine. 158 моделей · Лучший: 88.1% MMLU-ProАктивен Harder successor to MMLU: about 12,000 reasoning-heavy questions in 14 disciplines with 10 answer options. 243 моделей · Лучший: 91.2% Humanity's Last ExamАктивен About 2,500 expert-written questions across 100+ subjects from the Center for AI Safety and Scale AI, filtered to stump frontier models at creation time. 46 моделей · Лучший: 54.8%
Агентное программирование
SWE-bench VerifiedСпорный 500 human-validated GitHub issues from 12 Python repositories; the system must produce a patch that makes the hidden tests pass. 81 моделей · Лучший: 83.5% SWE-bench ProТолько ссылка Scale AI's harder, longer-horizon successor to SWE-bench with a 731-task public split and private/commercial splits. Только ссылка Terminal-Bench 2.0Активен Hard, realistic tasks an agent must finish in a sandboxed terminal (Stanford × Laude Institute). 45 моделей · Лучший: 84.7%
Математика
AIME (OTIS Mock AIME 2024–2025)Активен AIME-style competition maths (integer answers 0–999). We use Epoch AI's runs on the OTIS Mock AIME 2024–2025 papers, which are less exposed than past official AIME papers. 200 моделей · Лучший: 100.0% MATH Level 5Насыщен The hardest (level 5) problems of the MATH competition dataset (Hendrycks et al., 2021). 94 моделей · Лучший: 98.1%
Абстрактное мышление
ARC-AGI-1Насыщен Abstract grid puzzles (François Chollet) that test learning a new rule from a few examples. 6 моделей · Лучший: 98.5% ARC-AGI-2Активен Harder second version of ARC-AGI (2025), designed so that brute-force search is not enough. 4 моделей · Лучший: 95.0% ARC-AGI-3Активен Interactive, game-like third version of ARC-AGI that tests exploration and learning inside novel environments. 3 моделей · Лучший: 99.9%
Работа с компьютером
OSWorld (Verified)Активен 369 real computer tasks in a full Ubuntu desktop (browsers, office apps, file manager, terminal) checked by execution-based scripts. 9 моделей · Лучший: 72.1% OSWorld 2.0Активен Long-horizon successor to OSWorld (2026) with multi-application workflows. 9 моделей · Лучший: 31.4%
Инструменты и агенты
τ²-benchАктивен Sierra's customer-service agent benchmark: the agent follows a policy and uses tools while an LLM-simulated user talks to it (retail, airline, telecom, banking). 43 моделей · Лучший: 85.3% BFCL v4 (function calling)Активен Berkeley Function Calling Leaderboard: single, parallel and multi-turn tool calls, web search, memory and relevance detection. 80 моделей · Лучший: 77.5%
Китайский
CMMLUДавно не обновлялся Chinese multitask knowledge benchmark: 67 subjects including China-specific topics. 12 моделей · Лучший: 90.3% C-EvalДавно не обновлялся Chinese evaluation suite of about 13,900 multiple-choice questions over 52 disciplines and four difficulty levels. 11 моделей · Лучший: 68.7% SuperCLUEТолько ссылка Periodic Chinese general-capability evaluation of domestic and international models, published as reports. Только ссылка
Как мы собираем данные
- We read official data files and public datasets (Hugging Face, GitHub, CSV downloads) and never invent or estimate scores.
- Model names from different leaderboards are merged into one model page (for example dated versions and reasoning-effort variants).
- Leaderboards that restrict reuse are cited with their top entries only, or linked without copying.
- Data is refreshed by a repeatable script; each run keeps a dated snapshot so changes can be compared over time.
Тип источника
- Официальный рейтинг
- The benchmark maintainers' own leaderboard
- Независимый тест
- An independent evaluator ran the model itself (e.g. Epoch AI)
- Данные разработчика
- Self-reported or submitted by the model developer
- Статья / техотчёт
- Reported in a paper or technical report (usually by the developer)
- Сторонняя сводка
- Compiled by Epoch AI; the original source is not given per row
Источники данных и лицензии
| Источник | Лицензия / условия | Как мы используем | Последний сбор |
|---|---|---|---|
| LMArena leaderboard dataset LMArena, Arena Leaderboard Dataset (Hugging Face), CC BY 4.0 | CC BY 4.0 | Open licence – full data with attribution | 10 окт. 2026 г. ⚠︎ |
| Epoch AI Benchmarking Hub Epoch AI, 'Capabilities & benchmarking', https://epoch.ai/benchmarks (CC BY 4.0) | CC BY 4.0 | Open licence – full data with attribution | 22 окт. 2026 г. |
| SWE-bench leaderboards (swebench.com) SWE-bench team, swebench.com leaderboards | No explicit licence (NOASSERTION) – factual scores only, with citation | No explicit licence – factual scores only, with citation | 22 окт. 2026 г. |
| LiveCodeBench leaderboard LiveCodeBench (Jain et al.), livecodebench.github.io | No explicit licence on the site repo (benchmark code MIT) – factual scores only, with citation | No explicit licence – factual scores only, with citation | 22 окт. 2026 г. |
| EvalPlus leaderboard EvalPlus team, evalplus.github.io | Apache-2.0 (site repo) | Open licence – full data with attribution | 22 окт. 2026 г. |
| Berkeley Function Calling Leaderboard (BFCL) Gorilla / BFCL team, UC Berkeley | Apache-2.0 (gorilla repo) | Open licence – full data with attribution | 22 окт. 2026 г. |
| τ²-bench leaderboard submissions Sierra Research, τ²-bench | MIT (repo) | Open licence – full data with attribution | 22 окт. 2026 г. |
| MMLU-Pro leaderboard (TIGER-Lab) TIGER-Lab, MMLU-Pro leaderboard | Apache-2.0 (dataset) | Open licence – full data with attribution | 22 окт. 2026 г. |
| Hugging Face Open LLM Leaderboard v2 (archived) Hugging Face, Open LLM Leaderboard (archived 2025) | No licence tag – factual scores only, with citation | No explicit licence – factual scores only, with citation 4576 rows in archive; 353 official-developer models kept | 22 окт. 2026 г. |
| ARC Prize leaderboard ARC Prize Foundation, arcprize.org/leaderboard | ARC Prize Terms: no copying/aggregation for commercial purposes without permission | Restricted – only top entries cited top 10 per set only (terms) | 22 окт. 2026 г. |
| CMMLU leaderboard (GitHub README) Li et al., CMMLU (CC BY-NC-SA 4.0) | CC BY-NC-SA 4.0 | Restricted – only top entries cited | 22 окт. 2026 г. |
| C-Eval (GitHub README, initial release table) Huang et al., C-Eval | Code MIT; dataset CC BY-NC-SA 4.0 | Restricted – only top entries cited | 22 окт. 2026 г. |
| Artificial Analysis Artificial Analysis | Free API requires an API key and attribution; no open bulk data | Not collected – link only | — |
| SuperCLUE CLUE / SuperCLUE | No open data interface | Not collected – link only | — |
| Scale AI SEAL leaderboards (SWE-bench Pro, HLE) Scale AI | No open data interface | Not collected – link only | — |
Для ИИ-агентов
The same data as JSON: benchmark metadata with licences, every model's best scores and per-benchmark result lists with source links.
/api/benchmarks.json/api/benchmarks/matrix.json/api/benchmarks/<id>.json/llms.txt