Claude Opus 4 ranks first on 3 of the 104 benchmarks it has scores on.
Closed weights Flagship
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMLU-Pro | 87.3% | #12 of 212 | Official leaderboard | |
| SuperGPQA | 61.4% | #8 of 58 | Official leaderboard | |
| RewardBench 2 | 76.5% | #6 of 42 | Official leaderboard | |
| Lech Mazur Creative Writing | 8.36 | #7 of 44 | Epoch AI | |
| FaithJudge | 14.1% | #9 of 37 | Official leaderboard | |
| MMMLU | 88.8% | #17 of 62 | Model card | |
| ECLeKTic | 24% | #10 of 35 | Official leaderboard | |
| LMArena (Elo) | 1,426 | #96 of 333 | Official leaderboard | |
| WritingBench | 75.1 | #16 of 45 | Official leaderboard | |
| EQ-Bench 3 | 1,440 | #30 of 78 | Official leaderboard | |
| Vectara Hallucination Leaderboard (HHEM) | 4.80% | #59 of 148 | Official leaderboard | |
| BullshitBench | 32.7% | #45 of 107 | Official leaderboard | |
| Creative Writing v3 | 70.0 | #55 of 120 | Official leaderboard | |
| LLM Confabulation Leaderboard | 15.9% | #27 of 58 | Official leaderboard | |
| MultiNRC | 33.9% | #19 of 38 | Official leaderboard | |
| FACTS Grounding | 77.7% | #17 of 34 | Official leaderboard | |
| BullshitBench V2 | 34% | #66 of 122 | Official leaderboard | |
| MultiChallenge | 58.6% | #19 of 35 | Official leaderboard | |
| Fiction.LiveBench | 61.1% | #33 of 57 | Epoch AI | |
| Humanity's Last Exam | 10.7% | #29 of 47 | Official leaderboard | |
| Vectara Hallucination Leaderboard (2025 dataset) | 12% | #78 of 104 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Misguided Attention | 64.1% | #4 of 45 | Official leaderboard | |
| MATH Level 5 | 85% | #22 of 94 | Epoch AI | |
| AIME 2025 | 90% | #49 of 143 | Model card | |
| SimpleBench | 58.8% | #40 of 100 | Official leaderboard | |
| GPQA Diamond | 76.3% | #98 of 219 | Epoch AI | |
| DTBench | 69.3% | #82 of 178 | Official leaderboard | |
| FrontierMath Tier 4 | 4.17% | #28 of 55 | Epoch AI | |
| OTIS Mock AIME 2024-2025 | 64.4% | #111 of 202 | Epoch AI | |
| ARC-AGI-2 | 8.61% | #58 of 95 | Official leaderboard | |
| CritPt | 0.30% | #105 of 160 | Epoch AI | |
| FrontierMath | 4.48% | #48 of 70 | Epoch AI | |
| HMMT 2025 | 60% | #37 of 53 | Official leaderboard | |
| ForecastBench | 64.1 | #62 of 86 | Official leaderboard | |
| ARC-AGI | 35.7% | #71 of 92 | Official leaderboard | |
| AMO-Bench | 10.6% | #17 of 21 | Official leaderboard | |
| BRUMO 2025 | 81.7% | #35 of 42 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AutoCodeBench | 52.4% | #1 of 34 | Official leaderboard | |
| Aider Polyglot | 72% | #11 of 54 | Epoch AI | |
| METR Time Horizon | 84 min | #8 of 31 | Official leaderboard | |
| METR data deduplication | 32.6% | #1 of 3 | Model card | |
| METR Time Horizon 1.1 | 100 min | #13 of 26 | Official leaderboard | |
| WeirdML | 43.7% | #63 of 112 | Official leaderboard | |
| Internal model use survey (Anthropic) | 0% | #3 of 5 | Model card | |
| GSO-Bench | 6.86% | #19 of 30 | Official leaderboard | |
| LiveCodeBench v5 | 59.8% | #14 of 21 | Official leaderboard | |
| LiveCodeBench v6 | 56.6% | #14 of 21 | Official leaderboard | |
| SWE-bench Verified (hard subset) | 39.5% | #6 of 9 | Model card | |
| SWE-bench Verified | 70.7% | #24 of 32 | Epoch AI | |
| Quadruped reinforcement learning task (Anthropic) | 1.25 | #6 of 8 | Model card | |
| Text-based reinforcement learning task (Anthropic) | 0.63 | #6 of 7 | Model card | |
| Time series forecasting task (Anthropic) | 6.15 | #8 of 9 | Model card | |
| AlgoTune | 1.33× | #17 of 18 | Official leaderboard | |
| Internal AI research evaluation suite 2 | 0.35 | #6 of 6 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| tau-bench | 70.5% | #1 of 23 | Model card | |
| Terminal-Bench | 50% | #1 of 22 | Model card | |
| MCP-Universe | 28.1% | #9 of 30 | Official leaderboard | |
| Elimination Game | 4 | #23 of 56 | Official leaderboard | |
| LisanBench | 598.0 | #64 of 133 | Official leaderboard | |
| Step Race | 2 | #29 of 59 | Official leaderboard | |
| DeepResearch Bench | 46.8% | #12 of 24 | Epoch AI | |
| LLM Chess | 29 | #63 of 124 | Official leaderboard | |
| LMArena Search Arena | 1,152 | #33 of 35 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Vals Finance Agent v1 | 43.6% | #14 of 47 | Official leaderboard | |
| LegalBench | 83.1% | #60 of 138 | Official leaderboard | |
| TaxEval v2 | 71.9% | #68 of 129 | Official leaderboard | |
| MortgageTax | 58.6% | #65 of 86 | Official leaderboard | |
| Vending-Bench | $2,077 | #7 of 9 | Official leaderboard | |
| TaxCalcBench | 27.4% | #14 of 17 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMMU | 76.5% | #17 of 136 | Official leaderboard | |
| VISTA | 47% | #19 of 54 | Official leaderboard | |
| EnigmaEval | 5.57% | #22 of 41 | Epoch AI | |
| Vision Arena | 1,208 | #74 of 136 | Official leaderboard | |
| ZeroBench | 6.00% | #37 of 66 | Official leaderboard | |
| VPCT | 38% | #20 of 28 | Epoch AI | |
| Drone-Bench | 41% | #14 of 19 | Official leaderboard | |
| Blueprint-Bench | 0.35 | #6 of 8 | Official leaderboard | |
| GeoBench | 49% | #43 of 48 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HELM Capabilities | 78.0 | #8 of 60 | Official leaderboard | |
| LMCA | 37.4% | #62 of 139 | Epoch AI | |
| ACCoRD | 65.8 | #68 of 150 | Official leaderboard | |
| Conceptual Reasoning Index | 49.5 | #69 of 149 | Official leaderboard |
Anthropic's models in color, against every other model on the Epoch Capabilities Index, by release date.
The benchmarks the lab chose to headline, with the scores it gave.
Every result in the longer documents the lab published with it. 48 of them aren't in the launch post.