LMArena · GPQA · SWE-bench · HLE · ARC-AGI · OSWorld · BFCL
AI-modelbenchmarks
3086 scores voor 1079 modellen op 23 publieke benchmarks. Elk getal linkt naar de bron; opgaven van makers zijn apart gemarkeerd.
3086 scores 1079 modellen 23 benchmarks Bijgewerkt op 22 okt 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 Menselijke voorkeur GPQA Diamond Gemini 4 Argon 96.4% Wetenschap Humanity's Last Exam GPT-6 Astra 54.8% Kennis AIME (OTIS Mock AIME 2024–2025) Gemini 4 Argon 100.0% Wiskunde SWE-bench Verified Claude Opus 4.7 83.5% Agentisch programmeren Terminal-Bench 2.0 GPT-5.5 84.7% Agentisch programmeren OSWorld (Verified) Claude Sonnet 4.6 72.1% Computergebruik τ²-bench Gemini 3.0 Pro 85.3% Tools & agents BFCL v4 (function calling) Claude Opus 4.5 77.5% Tools & agents MMLU-Pro Gemini 3.1 Pro 91.2% Kennis
Matrix model × benchmark
Elke cel is de beste score in de hoofdinstelling (officiële ranglijsten en onafhankelijke tests eerst). ◇ = alleen opgave van maker of paper. Klik op een kolom om te sorteren, vink modellen aan om te vergelijken.
| Model | 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% | — | — | — | — | — | — |
Alle benchmarks
Programmeren
LMArena WebDevActief Arena where two models each build a web app from the same prompt and users vote on the result. 127 modellen · Beste: 1813 LiveCodeBenchNiet recent bijgewerkt 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 modellen · Beste: 75.8% HumanEval / HumanEval+Verzadigd 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 modellen · Beste: 89.0%
Algemeen
Open LLM Leaderboard v2Gearchiveerd 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 modellen · Beste: 48.0% Artificial Analysis Intelligence IndexAlleen link Independent composite index that runs a battery of evaluations plus price and speed measurements across API models. Alleen link
Kennis
MMLUVerzadigd About 14,000 multiple-choice questions over 57 subjects, from elementary maths to law and medicine. 158 modellen · Beste: 88.1% MMLU-ProActief Harder successor to MMLU: about 12,000 reasoning-heavy questions in 14 disciplines with 10 answer options. 243 modellen · Beste: 91.2% Humanity's Last ExamActief 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 modellen · Beste: 54.8%
Agentisch programmeren
SWE-bench VerifiedOmstreden 500 human-validated GitHub issues from 12 Python repositories; the system must produce a patch that makes the hidden tests pass. 81 modellen · Beste: 83.5% SWE-bench ProAlleen link Scale AI's harder, longer-horizon successor to SWE-bench with a 731-task public split and private/commercial splits. Alleen link Terminal-Bench 2.0Actief Hard, realistic tasks an agent must finish in a sandboxed terminal (Stanford × Laude Institute). 45 modellen · Beste: 84.7%
Wiskunde
AIME (OTIS Mock AIME 2024–2025)Actief 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 modellen · Beste: 100.0% MATH Level 5Verzadigd The hardest (level 5) problems of the MATH competition dataset (Hendrycks et al., 2021). 94 modellen · Beste: 98.1%
Abstract redeneren
ARC-AGI-1Verzadigd Abstract grid puzzles (François Chollet) that test learning a new rule from a few examples. 6 modellen · Beste: 98.5% ARC-AGI-2Actief Harder second version of ARC-AGI (2025), designed so that brute-force search is not enough. 4 modellen · Beste: 95.0% ARC-AGI-3Actief Interactive, game-like third version of ARC-AGI that tests exploration and learning inside novel environments. 3 modellen · Beste: 99.9%
Computergebruik
OSWorld (Verified)Actief 369 real computer tasks in a full Ubuntu desktop (browsers, office apps, file manager, terminal) checked by execution-based scripts. 9 modellen · Beste: 72.1% OSWorld 2.0Actief Long-horizon successor to OSWorld (2026) with multi-application workflows. 9 modellen · Beste: 31.4%
Tools & agents
τ²-benchActief 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 modellen · Beste: 85.3% BFCL v4 (function calling)Actief Berkeley Function Calling Leaderboard: single, parallel and multi-turn tool calls, web search, memory and relevance detection. 80 modellen · Beste: 77.5%
Chinees
CMMLUNiet recent bijgewerkt Chinese multitask knowledge benchmark: 67 subjects including China-specific topics. 12 modellen · Beste: 90.3% C-EvalNiet recent bijgewerkt Chinese evaluation suite of about 13,900 multiple-choice questions over 52 disciplines and four difficulty levels. 11 modellen · Beste: 68.7% SuperCLUEAlleen link Periodic Chinese general-capability evaluation of domestic and international models, published as reports. Alleen link
Werkwijze
- 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.
Brontype
- Officiële ranglijst
- The benchmark maintainers' own leaderboard
- Onafhankelijke test
- An independent evaluator ran the model itself (e.g. Epoch AI)
- Opgave van maker
- Self-reported or submitted by the model developer
- Paper / rapport
- Reported in a paper or technical report (usually by the developer)
- Verzameling door derden
- Compiled by Epoch AI; the original source is not given per row
Bronnen & licenties
| Bron | Licentie / voorwaarden | Ons gebruik | Laatst opgehaald |
|---|---|---|---|
| LMArena leaderboard dataset LMArena, Arena Leaderboard Dataset (Hugging Face), CC BY 4.0 | CC BY 4.0 | Open licence – full data with attribution | 10 okt 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 okt 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 okt 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 okt 2026 |
| EvalPlus leaderboard EvalPlus team, evalplus.github.io | Apache-2.0 (site repo) | Open licence – full data with attribution | 22 okt 2026 |
| Berkeley Function Calling Leaderboard (BFCL) Gorilla / BFCL team, UC Berkeley | Apache-2.0 (gorilla repo) | Open licence – full data with attribution | 22 okt 2026 |
| τ²-bench leaderboard submissions Sierra Research, τ²-bench | MIT (repo) | Open licence – full data with attribution | 22 okt 2026 |
| MMLU-Pro leaderboard (TIGER-Lab) TIGER-Lab, MMLU-Pro leaderboard | Apache-2.0 (dataset) | Open licence – full data with attribution | 22 okt 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 okt 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 okt 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 okt 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 okt 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 | — |
Voor AI-agents
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