LMArena · GPQA · SWE-bench · HLE · ARC-AGI · OSWorld · BFCL
Đánh giá mô hình AI
3086 điểm số của 1079 mô hình trên 23 benchmark công khai. Mỗi con số đều dẫn tới nguồn; số liệu do nhà phát triển tự công bố được đánh dấu riêng.
3086 điểm số 1079 mô hình 23 benchmark Cập nhật 22 thg 10, 2026
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 Ưu tiên của con người GPQA Diamond Gemini 4 Argon 96.4% Khoa học Humanity's Last Exam GPT-6 Astra 54.8% Kiến thức AIME (OTIS Mock AIME 2024–2025) Gemini 4 Argon 100.0% Toán học SWE-bench Verified Claude Opus 4.7 83.5% Lập trình bằng agent Terminal-Bench 2.0 GPT-5.5 84.7% Lập trình bằng agent OSWorld (Verified) Claude Sonnet 4.6 72.1% Sử dụng máy tính τ²-bench Gemini 3.0 Pro 85.3% Công cụ & agent BFCL v4 (function calling) Claude Opus 4.5 77.5% Công cụ & agent MMLU-Pro Gemini 3.1 Pro 91.2% Kiến thức
Ma trận mô hình × benchmark
Mỗi ô là điểm tốt nhất ở thiết lập chính (ưu tiên bảng xếp hạng chính thức và đánh giá độc lập). ◇ = chỉ có số liệu của nhà phát triển hoặc bài báo. Bấm vào cột để sắp xếp, đánh dấu mô hình để so sánh.
| Mô hình | 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% | — | — | — | — | — | — |
Tất cả benchmark
Lập trình
LMArena WebDevĐang cập nhật Arena where two models each build a web app from the same prompt and users vote on the result. 127 mô hình · Cao nhất: 1813 LiveCodeBenchLâu không cập nhật 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 mô hình · Cao nhất: 75.8% HumanEval / HumanEval+Bão hòa 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 mô hình · Cao nhất: 89.0%
Tổng hợp
Open LLM Leaderboard v2Đã lưu trữ 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 mô hình · Cao nhất: 48.0% Artificial Analysis Intelligence IndexChỉ liên kết Independent composite index that runs a battery of evaluations plus price and speed measurements across API models. Chỉ liên kết
Kiến thức
MMLUBão hòa About 14,000 multiple-choice questions over 57 subjects, from elementary maths to law and medicine. 158 mô hình · Cao nhất: 88.1% MMLU-ProĐang cập nhật Harder successor to MMLU: about 12,000 reasoning-heavy questions in 14 disciplines with 10 answer options. 243 mô hình · Cao nhất: 91.2% Humanity's Last ExamĐang cập nhật 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 mô hình · Cao nhất: 54.8%
Lập trình bằng agent
SWE-bench VerifiedGây tranh cãi 500 human-validated GitHub issues from 12 Python repositories; the system must produce a patch that makes the hidden tests pass. 81 mô hình · Cao nhất: 83.5% SWE-bench ProChỉ liên kết Scale AI's harder, longer-horizon successor to SWE-bench with a 731-task public split and private/commercial splits. Chỉ liên kết Terminal-Bench 2.0Đang cập nhật Hard, realistic tasks an agent must finish in a sandboxed terminal (Stanford × Laude Institute). 45 mô hình · Cao nhất: 84.7%
Toán học
AIME (OTIS Mock AIME 2024–2025)Đang cập nhật 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 mô hình · Cao nhất: 100.0% MATH Level 5Bão hòa The hardest (level 5) problems of the MATH competition dataset (Hendrycks et al., 2021). 94 mô hình · Cao nhất: 98.1%
Suy luận trừu tượng
ARC-AGI-1Bão hòa Abstract grid puzzles (François Chollet) that test learning a new rule from a few examples. 6 mô hình · Cao nhất: 98.5% ARC-AGI-2Đang cập nhật Harder second version of ARC-AGI (2025), designed so that brute-force search is not enough. 4 mô hình · Cao nhất: 95.0% ARC-AGI-3Đang cập nhật Interactive, game-like third version of ARC-AGI that tests exploration and learning inside novel environments. 3 mô hình · Cao nhất: 99.9%
Sử dụng máy tính
OSWorld (Verified)Đang cập nhật 369 real computer tasks in a full Ubuntu desktop (browsers, office apps, file manager, terminal) checked by execution-based scripts. 9 mô hình · Cao nhất: 72.1% OSWorld 2.0Đang cập nhật Long-horizon successor to OSWorld (2026) with multi-application workflows. 9 mô hình · Cao nhất: 31.4%
Công cụ & agent
τ²-benchĐang cập nhật 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 mô hình · Cao nhất: 85.3% BFCL v4 (function calling)Đang cập nhật Berkeley Function Calling Leaderboard: single, parallel and multi-turn tool calls, web search, memory and relevance detection. 80 mô hình · Cao nhất: 77.5%
Tiếng Trung
CMMLULâu không cập nhật Chinese multitask knowledge benchmark: 67 subjects including China-specific topics. 12 mô hình · Cao nhất: 90.3% C-EvalLâu không cập nhật Chinese evaluation suite of about 13,900 multiple-choice questions over 52 disciplines and four difficulty levels. 11 mô hình · Cao nhất: 68.7% SuperCLUEChỉ liên kết Periodic Chinese general-capability evaluation of domestic and international models, published as reports. Chỉ liên kết
Cách thu thập
- 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.
Loại nguồn
- Bảng xếp hạng chính thức
- The benchmark maintainers' own leaderboard
- Đánh giá độc lập
- An independent evaluator ran the model itself (e.g. Epoch AI)
- Nhà phát triển tự công bố
- Self-reported or submitted by the model developer
- Bài báo / báo cáo kỹ thuật
- Reported in a paper or technical report (usually by the developer)
- Tổng hợp của bên thứ ba
- Compiled by Epoch AI; the original source is not given per row
Nguồn dữ liệu & giấy phép
| Nguồn | Giấy phép / điều khoản | Cách chúng tôi sử dụng | Lần lấy gần nhất |
|---|---|---|---|
| LMArena leaderboard dataset LMArena, Arena Leaderboard Dataset (Hugging Face), CC BY 4.0 | CC BY 4.0 | Open licence – full data with attribution | 10 thg 10, 2026 ⚠︎ |
| 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 thg 10, 2026 |
| 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 thg 10, 2026 |
| 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 thg 10, 2026 |
| EvalPlus leaderboard EvalPlus team, evalplus.github.io | Apache-2.0 (site repo) | Open licence – full data with attribution | 22 thg 10, 2026 |
| Berkeley Function Calling Leaderboard (BFCL) Gorilla / BFCL team, UC Berkeley | Apache-2.0 (gorilla repo) | Open licence – full data with attribution | 22 thg 10, 2026 |
| τ²-bench leaderboard submissions Sierra Research, τ²-bench | MIT (repo) | Open licence – full data with attribution | 22 thg 10, 2026 |
| MMLU-Pro leaderboard (TIGER-Lab) TIGER-Lab, MMLU-Pro leaderboard | Apache-2.0 (dataset) | Open licence – full data with attribution | 22 thg 10, 2026 |
| 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 thg 10, 2026 |
| 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 thg 10, 2026 |
| 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 thg 10, 2026 |
| 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 thg 10, 2026 |
| 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 | — |
Dành cho AI agent
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