- Agentic terminal coding Terminal-Bench 2.0 65.4
- Agentic coding SWE-bench Verified 80.8
- Agentic computer use OSWorld 72.7
- Mean of Agentic tool use τ2-bench Retail / Agentic tool use τ2-bench Telecom 95.6
- Scaled tool use MCP Atlas 59.5
- Agentic search BrowseComp 84
- Multidisciplinary reasoning Humanity's Last Exam (with tools) 53
- Agentic financial analysis Finance Agent 60.7
- Office tasks GDPVal-AA Elo 1606
- Novel problem-solving ARC AGI 2 68.8
- Graduate-level reasoning GPQA Diamond 91.3
- Visual reasoning MMMU Pro (with tools) 77.3
- Multilingual Q&A MMMLU 91.1
Claude Opus 4.6
Claude Opus 4.6 ranks first on 16 of the 229 benchmarks it has scores on.
243
Benchmarks with a score
16
Ranked first, out of five or more models
53
In the top three
Anthropic Feb 2026
Claude Opus 4.6
Closed weights Flagship
155.8
#36 of 271
Epoch Capabilities Index
- Humanity's Last Exam 34.4% #11 /47
- SWE-bench Verified 78.7% #4 /32
- SWE-Bench Pro 51.9% #4 /24
Where it ranks
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.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMLU-Pro | 89.1% | #4 of 212 | Official leaderboard | |
| LMArena (Elo) | 1,503 | #7 of 333 | Official leaderboard | |
| Judgemark v4 | 90.7 | #1 of 40 | Official leaderboard | |
| BullshitBench | 92.7% | #3 of 107 | Official leaderboard | |
| BullshitBench V2 | 87% | #5 of 122 | Official leaderboard | |
| Judgemark v2.1 | 92.9 | #2 of 42 | Official leaderboard | |
| EQ-Bench 3 | 1,712 | #4 of 78 | Official leaderboard | |
| Document Arena | 1,495 | #2 of 37 | Official leaderboard | |
| MMMLU | 91.1% | #6 of 62 | Model card | |
| Phare | 74.1% | #9 of 62 | Official leaderboard | |
| Creative Writing v3 | 78.3 | #20 of 120 | Official leaderboard | |
| Extended NYT Connections | 92.1% | #17 of 97 | Official leaderboard | |
| Longform Creative Writing (EQ-Bench) | 77.7 | #24 of 119 | Official leaderboard | |
| MultiNRC | 57.1% | #8 of 38 | Official leaderboard | |
| Humanity's Last Exam | 34.4% | #11 of 47 | Official leaderboard | |
| CL-bench Life | 17% | #4 of 16 | Official leaderboard | |
| FACTS Search | 76.7% | #8 of 31 | Official leaderboard | |
| CL-bench | 20.7% | #7 of 24 | Official leaderboard | |
| SimpleQA Verified | 45.2% | #18 of 56 | Official leaderboard | |
| ECLeKTic | 37.3% | #4 of 10 | Model card | |
| Honesty human feedback win rate (Anthropic) | 61% | #3 of 6 | Model card | |
| SkateBench v2 | 64.4% | #11 of 21 | Official leaderboard | |
| EQ-Bench 4 | 1,210 | #17 of 32 | Official leaderboard | |
| FACTS Multimodal | 36.8% | #24 of 39 | Official leaderboard | |
| MultiChallenge | 56% | #23 of 35 | Official leaderboard | |
| AA-Omniscience | 20.8% | #12 of 17 | Model card | |
| 100Q-Hard (Anthropic) | 21.5% | #5 of 7 | Model card | |
| FACTS Benchmark Suite | 36.8% | #31 of 40 | Official leaderboard | |
| Vectara Hallucination Leaderboard (2025 dataset) | 12.2% | #83 of 104 | Official leaderboard | |
| False-premise questions (Anthropic) | 96.8% | #6 of 6 | Model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AIME 2025 | 99.8% | #9 of 143 | Model card | |
| FrontierMath | 40.7% | #8 of 70 | Epoch AI | |
| HMMT 2026 | 96.2% | #5 of 31 | Official leaderboard | |
| FrontierMath Tier 4 | 22.9% | #9 of 55 | Epoch AI | |
| MathArena (Apex, IMO, USAMO) | 34.5% | #8 of 47 | Official leaderboard | |
| Echo General AI Prediction Leaderboard | 1,009 | #3 of 17 | Official leaderboard | |
| GPQA Diamond | 90.5% | #39 of 219 | Epoch AI | |
| ProofBench | 50% | #12 of 59 | Official leaderboard | |
| DTBench | 85.3% | #37 of 178 | Official leaderboard | |
| SimpleBench | 67.6% | #22 of 100 | Official leaderboard | |
| OTIS Mock AIME 2024-2025 | 94.4% | #45 of 202 | Epoch AI | |
| ARC-AGI | 94% | #23 of 92 | Official leaderboard | |
| AIME 2026 | 96.7% | #8 of 31 | Official leaderboard | |
| ResearchClawBench | 21.5 | #6 of 23 | Official leaderboard | |
| ARC-AGI-2 | 69.2% | #26 of 95 | Official leaderboard | |
| FrontierMath Tiers 1-3 v2 | 66% | #34 of 89 | Epoch AI | |
| NatureBench | 11.1% | #8 of 18 | Official leaderboard | |
| IMProofBench | 31.7% | #5 of 10 | Official leaderboard | |
| FrontierMath Tier 4 v2 | 26.8% | #38 of 71 | Epoch AI | |
| ForecastBench | 63.3 | #51 of 86 | Official leaderboard | |
| USAMO 2026 | 47% | #6 of 9 | Official leaderboard | |
| BioPipelineBench (Anthropic) | 53.1% | #2 of 3 | Model card | |
| Phylogenetics (Anthropic) | 61.3% | #2 of 3 | Model card | |
| Leipzig Benchmark | 14% | #14 of 19 | Official leaderboard | |
| LABBench2 | 51.9% | #4 of 5 | Official leaderboard | |
| Organic Chemistry (Anthropic) | 53.9% | #4 of 5 | Model card | |
| BioMysteryBench | 61.5% | #7 of 8 | Model card |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| METR Time Horizon | 12.0 h | #2 of 39 | Epoch AI | |
| LiveSQLBench | 38% | #2 of 36 | Official leaderboard | |
| METR Time Horizon 1.1 | 12.0 h | #2 of 26 | Official leaderboard | |
| WeirdML | 78% | #12 of 112 | Official leaderboard | |
| Vibe Code Bench v1 | 36.1% | #2 of 17 | Official leaderboard | |
| OpenHands Index | 66.7% | #4 of 33 | Official leaderboard | |
| SWE-bench Verified | 78.7% | #4 of 32 | Epoch AI | |
| Bird-SQL | 70.2% | #2 of 13 | Official leaderboard | |
| SWE-Bench Pro | 51.9% | #4 of 24 | Official leaderboard | |
| MLE-bench | 63.1% | #2 of 12 | Official leaderboard | |
| SWE-fficiency | 15.5% | #3 of 17 | Official leaderboard | |
| SWE-bench Verified (hard subset) | 47.2% | #2 of 9 | Model card | |
| EvoCode-Bench | 44% | #3 of 13 | Official leaderboard | |
| GSO-Bench | 41.2% | #7 of 30 | Official leaderboard | |
| OJBench | 67.2% | #2 of 8 | Official leaderboard | |
| WebDev Arena | 1,556 | #30 of 105 | Epoch AI | |
| SWE-bench Multilingual | 77.8% | #20 of 70 | Model card | |
| Text-based reinforcement learning task (Anthropic) | 0.97 | #2 of 7 | Model card | |
| PostTrainBench | 24.8% | #7 of 23 | Official leaderboard | |
| SWE-bench-Live | 63% | #4 of 13 | Official leaderboard | |
| Code Arena | 1,546 | #41 of 124 | Official leaderboard | |
| Internal AI research evaluation suite 2 | 0.61 | #2 of 6 | Model card | |
| SlopCodeBench | 20.9% | #5 of 14 | Official leaderboard | |
| ALE-Bench | 997 | #55 of 121 | Epoch AI | |
| Vibe Code Bench | 57.6% | #56 of 115 | Official leaderboard | |
| OpenRCA | 34.9% | #1 of 2 | Model card | |
| Image-to-WebDev Arena | 1,543 | #28 of 55 | Official leaderboard | |
| GBAEval | 44.1% | #11 of 21 | Official leaderboard | |
| TasteVal | 40.4× | #11 of 20 | Official leaderboard | |
| ProgramBench | 0% | #14 of 25 | Official leaderboard | |
| SWE Atlas | 33.3% | #13 of 22 | Official leaderboard | |
| SWE Atlas Test Writing | 36.7% | #14 of 22 | Official leaderboard | |
| SWE Atlas Refactoring | 35.6% | #11 of 17 | Official leaderboard | |
| AlgoTune | 1.47× | #12 of 18 | Official leaderboard | |
| MirrorCode | 33.3% | #19 of 28 | Epoch AI | |
| FrontierCode | 26.6% | #35 of 46 | Epoch AI | |
| FrontierCode 1.1 | 26.9% | #7 of 8 | Official leaderboard | |
| Internal model use survey (Anthropic) | 0% | #5 of 5 | Model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| DeepResearch Bench | 55.3% | #1 of 24 | Epoch AI | |
| Claw-Eval Pass^3 | 70.4% | #1 of 23 | Official leaderboard | |
| VitaBench 2.0 | 50.3% | #1 of 20 | Official leaderboard | |
| LisanBench | 14206.3 | #8 of 133 | Official leaderboard | |
| tau2-bench | 95.6% | #5 of 69 | Model card | |
| Claw-Eval-Live | 66.7% | #1 of 12 | Official leaderboard | |
| SnakeBench | 33 | #18 of 183 | Official leaderboard | |
| Terminal-Bench 2.0 | 76.4% | #5 of 43 | Official leaderboard | |
| WebArena | 68% | #1 of 8 | Model card | |
| Pencil Puzzle Bench | 33.3% | #8 of 53 | Official leaderboard | |
| ClawBench | 61.4% | #1 of 6 | Official leaderboard | |
| Claw-Eval | 80.4% | #4 of 23 | Official leaderboard | |
| Toolathlon | 47.2% | #11 of 50 | Official leaderboard | |
| MyPCBench | 58.2% | #2 of 9 | Official leaderboard | |
| PinchBench | 74.4% | #12 of 48 | Official leaderboard | |
| LMArena Search Arena | 1,222 | #9 of 35 | Official leaderboard | |
| BrowseComp | 84% | #20 of 77 | Model card | |
| DeepSearchQA | 91.3% | #6 of 22 | Model card | |
| iOSWorld | 51.9% | #1 of 3 | Official leaderboard | |
| Game Arena Word Association | 287 | #13 of 36 | Official leaderboard | |
| Game Arena Clobber | 326 | #15 of 36 | Official leaderboard | |
| LLM Chess | 198 | #53 of 124 | Official leaderboard | |
| Mystery Game Puzzles | 25% | #37 of 80 | Epoch AI | |
| MCP Atlas | 76.8% | #16 of 34 | Official leaderboard | |
| Game Arena Ultimate Tic-Tac-Toe | 286 | #17 of 36 | Official leaderboard | |
| SkillsBench | 50.2% | #10 of 21 | Official leaderboard | |
| Game Arena Lines of Action | 496 | #18 of 36 | Official leaderboard | |
| Game Arena Werewolf | -1 | #4 of 8 | Official leaderboard | |
| Game Arena Word Art | 451 | #18 of 35 | Official leaderboard | |
| OSWorld-Verified | 72.7% | #20 of 37 | Model card | |
| Game Arena Bargaining | 178 | #20 of 36 | Official leaderboard | |
| PinchBench 2.0 | 69.9% | #32 of 57 | Official leaderboard | |
| Game Arena Checkers | 885 | #22 of 38 | Official leaderboard | |
| Kaggle Game Arena | 259 | #23 of 39 | Official leaderboard | |
| Game Arena Dots and Boxes | 872 | #22 of 37 | Official leaderboard | |
| Game Arena Coin Game | 310 | #21 of 35 | Official leaderboard | |
| τ³-bench Banking | 27.3% | #15 of 25 | Official leaderboard | |
| Game Arena Poker | -1.90 | #19 of 31 | Official leaderboard | |
| Game Arena Go | 568 | #23 of 36 | Official leaderboard | |
| Game Arena Reversi | 328 | #24 of 36 | Official leaderboard | |
| ARC-AGI-3 | 0.51% | #16 of 24 | Official leaderboard | |
| Game Arena Nine Men's Morris | 523 | #25 of 36 | Official leaderboard | |
| Game Arena Four in a Row | 210 | #27 of 38 | Official leaderboard | |
| Chess Puzzles | 21% | #59 of 83 | Epoch AI | |
| MobileWorld | 44.5% | #13 of 18 | Official leaderboard | |
| EBR-bench | 12.7% | #22 of 28 | Epoch AI | |
| Game Arena Dark Hex | 109 | #32 of 38 | Official leaderboard | |
| Game Arena Five in a Row | 315 | #17 of 20 | Official leaderboard | |
| ApprenticeBench | 5.00% | #17 of 18 | Official leaderboard | |
| Game Arena Chess | 50 | #38 of 39 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Vals Finance Agent v1 | 60.6% | #1 of 47 | Official leaderboard | |
| TaxEval v2 | 76% | #8 of 129 | Official leaderboard | |
| ProfBench | 58.8% | #5 of 72 | Official leaderboard | |
| Vals Finance Agent | 60% | #5 of 49 | Official leaderboard | |
| CorpFin v2 | 67% | #14 of 119 | Official leaderboard | |
| SpreadsheetBench 2 | 34.9% | #1 of 8 | Official leaderboard | |
| PRBench | 52.3% | #7 of 44 | Official leaderboard | |
| LegalBench | 85.3% | #22 of 138 | Official leaderboard | |
| MortgageTax | 68.5% | #15 of 86 | Official leaderboard | |
| TaxCalcBench | 52.9% | #3 of 17 | Official leaderboard | |
| MedScribe | 86.7% | #21 of 98 | Official leaderboard | |
| PhysicianBench | 31.7% | #3 of 13 | Official leaderboard | |
| Vending-Bench 2 | $8,018 | #17 of 66 | Official leaderboard | |
| MedCode | 49.1% | #26 of 96 | Official leaderboard | |
| FORTE | 73.2% | #6 of 21 | Official leaderboard | |
| EnterpriseClawBench | 63.5% | #3 of 9 | Official leaderboard | |
| GDPval-AA | 1,606 | #6 of 16 | Model card | |
| CaseLaw v2 | 62.1% | #20 of 53 | Official leaderboard | |
| E-Commerce Bench | 2.58 | #10 of 19 | Official leaderboard | |
| Vending-Bench Arena | $5,062 | #17 of 32 | Official leaderboard | |
| Remote Labor Index | 4.17% | #10 of 17 | Official leaderboard | |
| CADGenBench | 30.9% | #7 of 11 | Official leaderboard | |
| RCM-Bench-1 | 87.8% | #13 of 20 | Official leaderboard | |
| APEX-Agents | 46.3% | #37 of 56 | Epoch AI | |
| Real-World Finance (Anthropic) | 64.1% | #2 of 3 | Model card | |
| prinzbench | 19.0 | #24 of 26 | Official leaderboard | |
| FoodTruck Bench | $49,519 | #10 of 10 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Vision Arena | 1,299 | #3 of 136 | Official leaderboard | |
| Claw-Eval-MM | 54.7% | #1 of 11 | Official leaderboard | |
| MATH-Vision (MathVision) | 84.6% | #14 of 105 | Official leaderboard | |
| OCR Arena | 1,713 | #3 of 11 | Official leaderboard | |
| OCRBench v2 | 48.4% | #20 of 65 | Official leaderboard | |
| MMMU-Pro | 77.3% | #24 of 72 | Model card | |
| ZeroBench | 12% | #25 of 66 | Official leaderboard | |
| EnigmaEval | 7.60% | #17 of 41 | Epoch AI | |
| VISTA | 46.1% | #23 of 54 | Official leaderboard | |
| ParseBench | 54.1% | #30 of 42 | Official leaderboard | |
| CharXiv | 68.5% | #51 of 67 | Model card | |
| Furniture Assembly Benchmark | 28.3% | #27 of 35 | Epoch AI | |
| VoxelBench | 1,394 | #45 of 56 | Official leaderboard | |
| ClockBench | 10% | #30 of 37 | Official leaderboard | |
| Math Kangaroo 2025 (MathArena) | 72.3% | #18 of 22 | Official leaderboard | |
| BabyVision | 14.8% | #25 of 27 | Official leaderboard |
Safety and security
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| ACCoRD | 80.4 | #11 of 150 | Official leaderboard | |
| LMCA | 55.8% | #13 of 139 | Epoch AI | |
| Conceptual Reasoning Index | 66.6 | #15 of 149 | Official leaderboard | |
| SAGE (Vals) | 51.6% | #12 of 81 | Official leaderboard | |
| CAIS Risk Index | 43.2 | #20 of 65 | Official leaderboard | |
| CAIS Text Capabilities Index | 37.5 | #24 of 66 | Official leaderboard | |
| CAIS Vision Capabilities Index | 47.7 | #31 of 53 | Official leaderboard |
Anthropic's models over time
Anthropic's models in color, against every other model on the Epoch Capabilities Index, by release date.
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.
Its system card
Every result in the longer documents the lab published with it. 80 of them aren't in the launch post.
System Card: Claude Opus 4.6 5 Feb 2026
Capabilities
- SWE-bench Verified 80.8
- Terminal-Bench 2.0 65.4
- tau2-bench
- MCP Atlas 59.5
- OSWorld-Verified 72.7
- ARC-AGI-2 68.8
- GPQA Diamond 91.3
- MMMU-Pro 73.9
- MMMLU 91.1
- SWE-bench Multilingual 77.8
- OpenRCA 34.9
- ARC-AGI 94
- GDPval-AA 1606
- AIME 2025 99.8
- Vals Finance Agent 60.7
- Real-World Finance (Anthropic) 64.1
- Vending-Bench 2 8017.6
- CyberGym 66.6
- OpenAI MRCR v2
- GraphWalks
- LAB-Bench FigQA 58
- CharXiv 68.5
- WebArena 68
- BrowseComp 83.7
- Humanity's Last Exam 40
- DeepSearchQA 91.3
- BioPipelineBench (Anthropic) 53.1
- BioMysteryBench 61.5
- Structural Biology (Anthropic)
- Organic Chemistry (Anthropic) 53.9
- Phylogenetics (Anthropic) 61.3
Hallucination
Safety
- Single-turn harmful request evaluations (Anthropic) 99.6
- Single-turn benign request evaluations (Anthropic) 0.68
- Higher-difficulty harmful request evaluations (Anthropic) 99.2
- Higher-difficulty benign request evaluations (Anthropic) 0.04
- Multi-turn testing (Anthropic)
- Child safety evaluations (Anthropic)
- Suicide and self-harm evaluations (Anthropic)
- SSH stress-testing evaluation (Anthropic) 95
- Agentic coding malicious requests (Anthropic) 99.3
- Malicious use of Claude Code (Anthropic) 83.2
- Malicious computer use (Anthropic) 88.3
Robustness
Alignment
- External pilot use failure modes (Anthropic)
- Reward-hack-prone coding tasks (Anthropic) 0
- Impossible tasks (Anthropic) 50
- Agentic code behavior scores (Anthropic)
- Over-eager behavior on GUI computer use tasks (Anthropic) 11.4
- Anthropic automated behavioral audit
- Petri
- UK AISI alignment and monitorability evaluations
- Refusal to assist with AI safety R&D (Anthropic) 14
- Self-preference bias (Anthropic)
- Internal codebase sabotage propensity (Anthropic) 0
- Participation in junk science (Anthropic) 1.1
- Deference to governments in local languages (Anthropic)
- SHADE-Arena 18
- Agentic misalignment suite (Anthropic)
Bio and chem risk
Cyber risk
AI R&D risk
- SWE-bench Verified (hard subset) 47.2
- Internal model use survey (Anthropic) 0
- Kernels task (Anthropic) 190
- Time series forecasting task (Anthropic)
- Text-based reinforcement learning task (Anthropic) 0.975
- LLM training task (Anthropic) 34
- Quadruped reinforcement learning task (Anthropic)
- Novel compiler task (Anthropic)
- Internal AI research evaluation suite 2 0.612