- Agentic coding SWE-bench Pro 64.3
- Agentic coding SWE-bench Verified 87.6
- SWE-bench Multilingual 80.5
- SWE-bench Multimodal 34.5
- Agentic terminal coding Terminal-Bench 2.0 69.4
- Multidisciplinary reasoning Humanity's Last Exam (with tools) 54.7
- Agentic search BrowseComp 79.3
- Scaled tool use MCP-Atlas 77.3
- Agentic computer use OSWorld-Verified 78
- Agentic financial analysis Finance Agent v1.1 64.4
- Graduate-level reasoning GPQA Diamond 94.2
- Visual reasoning CharXiv Reasoning (with tools) 91
- Multilingual Q&A MMMLU 91.5
- ScreenSpot-Pro (with tools) 87.6
- OfficeQA 86.3
- OfficeQA Pro 80.6
- ARC-AGI-1 92
- ARC-AGI-2 75.8
Claude Opus 4.7
Claude Opus 4.7 ranks first on 13 of the 217 benchmarks it has scores on.
229
Benchmarks with a score
13
Ranked first, out of five or more models
38
In the top three
Anthropic Apr 2026
Claude Opus 4.7
Closed weights Flagship
156.7
#29 of 271
Epoch Capabilities Index
- Humanity's Last Exam 36.2% #10 /47
- SWE-bench Verified 83.5% #1 /32
- Terminal-Bench 2.0 80.2% #3 /43
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 | |
|---|---|---|---|---|
| LMArena (Elo) | 1,501 | #8 of 333 | Official leaderboard | |
| HELM Arabic | 84.1 | #1 of 28 | Official leaderboard | |
| EQ-Bench 3 | 1,914 | #3 of 78 | Official leaderboard | |
| HELM Arabic Enterprise | 77.9 | #1 of 23 | Official leaderboard | |
| BullshitBench | 80% | #7 of 107 | Official leaderboard | |
| BullshitBench V2 | 83% | #8 of 122 | Official leaderboard | |
| Creative Writing v3 | 80.7 | #9 of 120 | Official leaderboard | |
| Judgemark v4 | 84.0 | #3 of 40 | Official leaderboard | |
| Longform Creative Writing (EQ-Bench) | 81.8 | #9 of 119 | Official leaderboard | |
| MMMLU | 91.5% | #5 of 62 | Model card | |
| Document Arena | 1,494 | #3 of 37 | Official leaderboard | |
| INCLUDE | 87% | #3 of 25 | Model card | |
| FACTS Search | 78.8% | #5 of 31 | Official leaderboard | |
| Humanity's Last Exam | 36.2% | #10 of 47 | Official leaderboard | |
| EQ-Bench 4 | 1,293 | #7 of 32 | Official leaderboard | |
| SimpleQA Verified | 51.7% | #26 of 84 | Epoch AI | |
| AA-Omniscience | 34.4% | #9 of 17 | Model card | |
| Global MMLU | 89.9% | #6 of 11 | Model card | |
| FACTS Parametric | 38.8% | #25 of 45 | Official leaderboard | |
| MILU | 89.9% | #6 of 10 | Model card | |
| FACTS Multimodal | 35.8% | #25 of 39 | Official leaderboard | |
| False-premise questions (Anthropic) | 77.2% | #4 of 6 | Model card | |
| ECLeKTic | 24.1% | #7 of 10 | Model card | |
| Extended NYT Connections | 39% | #71 of 97 | Official leaderboard | |
| Vectara Hallucination Leaderboard (2025 dataset) | 12% | #80 of 104 | Official leaderboard | |
| FACTS Benchmark Suite | 35.8% | #32 of 40 | Official leaderboard | |
| 100Q-Hard (Anthropic) | 17.1% | #7 of 7 | Model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| FrontierMath | 43.8% | #6 of 70 | Epoch AI | |
| FrontierMath Tier 4 | 22.9% | #8 of 55 | Epoch AI | |
| MathArena (Apex, IMO, USAMO) | 40.6% | #7 of 47 | Official leaderboard | |
| DTBench | 91.1% | #27 of 178 | Official leaderboard | |
| OTIS Mock AIME 2024-2025 | 97.8% | #31 of 202 | Epoch AI | |
| ProofBench | 54% | #10 of 59 | Official leaderboard | |
| GPQA Diamond | 90.2% | #42 of 219 | Epoch AI | |
| xbench-ScienceQA | 73% | #6 of 29 | Official leaderboard | |
| ARC-AGI-2 | 75.8% | #20 of 95 | Official leaderboard | |
| NatureBench | 15.6% | #4 of 18 | Official leaderboard | |
| ARC-AGI | 93.5% | #25 of 92 | Official leaderboard | |
| ResearchClawBench | 20.7 | #7 of 23 | Official leaderboard | |
| SimpleBench | 61.7% | #31 of 100 | Official leaderboard | |
| FrontierMath Tiers 1-3 v2 | 70.2% | #28 of 89 | Epoch AI | |
| HMMT 2026 | 93.9% | #10 of 31 | Official leaderboard | |
| CritPt | 12% | #53 of 160 | Epoch AI | |
| Phylogenetics (Anthropic) | 79.6% | #1 of 3 | Model card | |
| FrontierMath Tier 4 v2 | 31.7% | #31 of 71 | Epoch AI | |
| AIME 2026 | 95.8% | #15 of 31 | Official leaderboard | |
| ForecastBench | 62.9 | #44 of 86 | Official leaderboard | |
| Organic Chemistry (Anthropic) | 77.2% | #3 of 5 | Model card | |
| LatchBio Capabilities v1.0 | 38.6% | #18 of 27 | Official leaderboard | |
| Leipzig Benchmark | 13% | #15 of 19 | Official leaderboard | |
| USAMO 2026 | 69.3% | #6 of 6 | Model card | |
| BioPipelineBench Verified (Anthropic) | 83.6% | #1 of 1 | Model card |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SWE-bench Verified | 83.5% | #1 of 32 | Epoch AI | |
| EvoCode-Bench | 54% | #1 of 13 | Official leaderboard | |
| OpenHands Index | 69.7% | #3 of 33 | Official leaderboard | |
| WeirdML | 76.4% | #15 of 112 | Official leaderboard | |
| SWE-bench Multilingual | 80.5% | #12 of 70 | Model card | |
| PostTrainBench | 28.6% | #4 of 23 | Official leaderboard | |
| IOI (Vals) | 47.1% | #10 of 55 | Official leaderboard | |
| GSO-Bench | 44.1% | #6 of 30 | Official leaderboard | |
| SWE-Bench Pro | 64.3% | #16 of 77 | Model card | |
| Time series forecasting task (Anthropic) | 4.78 | #2 of 9 | Model card | |
| SWE Atlas Refactoring | 48.6% | #4 of 17 | Official leaderboard | |
| ALE-Bench | 1,323 | #30 of 121 | Epoch AI | |
| FrontierCode | 23% | #3 of 12 | Official leaderboard | |
| WebDev Arena | 1,567 | #28 of 105 | Epoch AI | |
| Code Arena | 1,558 | #38 of 124 | Official leaderboard | |
| Code Migration | 43.9% | #26 of 78 | Official leaderboard | |
| SWE-Marathon | 16% | #4 of 12 | Official leaderboard | |
| Vibe Code Bench | 71% | #43 of 115 | Official leaderboard | |
| Quadruped reinforcement learning task (Anthropic) | 26.5 | #3 of 8 | Model card | |
| Image-to-WebDev Arena | 1,577 | #22 of 55 | Official leaderboard | |
| TasteVal | 50.1× | #8 of 20 | Official leaderboard | |
| SlopCodeBench | 20.9% | #6 of 14 | Official leaderboard | |
| SWE Atlas | 40.3% | #11 of 22 | Official leaderboard | |
| FrontierCS | 43.0 | #7 of 14 | Official leaderboard | |
| AndroidBench | 68.7% | #17 of 32 | Official leaderboard | |
| GBAEval | 43.8% | #12 of 21 | Official leaderboard | |
| SWE Atlas Test Writing | 38.5% | #13 of 22 | Official leaderboard | |
| FrontierCode 1.1 | 38.5% | #5 of 8 | Official leaderboard | |
| PostTrainBench 1.1 | 28.6% | #7 of 11 | Official leaderboard | |
| ProgramBench | 0% | #19 of 25 | Official leaderboard | |
| Vals RSI Index | 0.19 | #19 of 25 | Official leaderboard | |
| PostTrainBench 1.2 | 24.3% | #13 of 17 | Official leaderboard | |
| Senior SWE-Bench | 18.1% | #18 of 23 | Official leaderboard | |
| MirrorCode | 31.1% | #22 of 28 | Epoch AI | |
| SWE-bench Multimodal | 34.5% | #8 of 10 | Model card | |
| Novel compiler task (Anthropic) | 71.1% | #4 of 4 | Model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LisanBench | 14400.0 | #7 of 133 | Official leaderboard | |
| OSWorld-Verified | 83.6% | #2 of 33 | Official leaderboard | |
| Terminal-Bench 2.0 | 80.2% | #3 of 43 | Official leaderboard | |
| Toolathlon | 52.8% | #4 of 50 | Official leaderboard | |
| Pencil Puzzle Bench | 50% | #5 of 53 | Official leaderboard | |
| SkillsBench | 61.2% | #2 of 21 | Official leaderboard | |
| DeepTerminalBench | 34% | #1 of 9 | Official leaderboard | |
| ScreenSpot-Pro | 87.6% | #3 of 25 | Model card | |
| ClawBench V2 | 44.6% | #1 of 7 | Official leaderboard | |
| SnakeBench | 30 | #37 of 183 | Official leaderboard | |
| τ³-bench Banking | 40.2% | #7 of 25 | Official leaderboard | |
| LLM Chess | 691 | #35 of 124 | Official leaderboard | |
| OSWorld 2.0 | 18.2% | #2 of 7 | Official leaderboard | |
| MobileWorld | 56.4% | #6 of 18 | Official leaderboard | |
| LMArena Search Arena | 1,210 | #12 of 35 | Official leaderboard | |
| BrowseComp | 79.3% | #28 of 77 | Model card | |
| PinchBench 2.0 | 76% | #21 of 57 | Official leaderboard | |
| MCP Atlas | 79.1% | #13 of 34 | Official leaderboard | |
| Chess Puzzles | 41% | #34 of 83 | Epoch AI | |
| CEO-Bench | $70,620 | #8 of 19 | Official leaderboard | |
| Mystery Game Puzzles | 28% | #34 of 80 | Epoch AI | |
| Game Arena Bargaining | 195 | #16 of 36 | Official leaderboard | |
| DeepSearchQA | 89.1% | #10 of 22 | Model card | |
| Game Arena Checkers | 943 | #18 of 38 | Official leaderboard | |
| Game Arena Word Association | 207 | #19 of 36 | Official leaderboard | |
| Game Arena Coin Game | 313 | #20 of 35 | Official leaderboard | |
| Game Arena Word Art | 325 | #21 of 35 | Official leaderboard | |
| Game Arena Nine Men's Morris | 542 | #23 of 36 | Official leaderboard | |
| EBR-bench | 19.1% | #18 of 28 | Epoch AI | |
| Kaggle Game Arena | 209 | #27 of 39 | Official leaderboard | |
| Game Arena Poker | -6.53 | #22 of 31 | Official leaderboard | |
| Game Arena Go | 508 | #27 of 36 | Official leaderboard | |
| Game Arena Five in a Row | 353 | #16 of 20 | Official leaderboard | |
| Game Arena Four in a Row | 193 | #31 of 38 | Official leaderboard | |
| DRACO | 77.7% | #9 of 11 | Model card | |
| ApprenticeBench | 7.00% | #15 of 18 | Official leaderboard | |
| Terminal-Bench 2.1 | 68.9% | #11 of 13 | Official leaderboard | |
| Game Arena Ultimate Tic-Tac-Toe | 57 | #31 of 36 | Official leaderboard | |
| Game Arena Dark Hex | 97 | #34 of 38 | Official leaderboard | |
| Game Arena Chess | 284 | #35 of 39 | Official leaderboard | |
| ARC-AGI-3 | 0.18% | #22 of 24 | Official leaderboard | |
| Game Arena Clobber | 25 | #34 of 36 | Official leaderboard | |
| Game Arena Reversi | 40 | #34 of 36 | Official leaderboard | |
| Game Arena Dots and Boxes | 387 | #36 of 37 | Official leaderboard | |
| Game Arena Lines of Action | 0 | #36 of 36 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Vals Finance Agent | 64.4% | #1 of 49 | Official leaderboard | |
| ProfBench | 61.3% | #2 of 72 | Official leaderboard | |
| MortgageTax | 70.3% | #3 of 86 | Official leaderboard | |
| OfficeQA Pro | 80.6% | #1 of 14 | Model card | |
| MedCode | 54.9% | #7 of 96 | Official leaderboard | |
| OfficeQA | 86.3% | #1 of 12 | Model card | |
| CaseLaw v2 | 68.4% | #5 of 53 | Official leaderboard | |
| FORTE | 77.6% | #2 of 21 | Official leaderboard | |
| Vending-Bench 2 | $10,937 | #8 of 66 | Official leaderboard | |
| TaxEval v2 | 75.3% | #21 of 129 | Official leaderboard | |
| LegalBench | 85.3% | #24 of 138 | Official leaderboard | |
| GDPval-AA | 1,753 | #3 of 16 | Model card | |
| CorpFin v2 | 66.1% | #23 of 119 | Official leaderboard | |
| CommerceAgentBench | 52.3% | #3 of 13 | Official leaderboard | |
| CADGenBench | 36.9% | #3 of 11 | Official leaderboard | |
| PhysicianBench | 29.3% | #4 of 13 | Official leaderboard | |
| Excel Modeling Benchmark | 63.7% | #27 of 76 | Official leaderboard | |
| Harvey LAB (Legal Agent Benchmark) | 6.67% | #30 of 80 | Official leaderboard | |
| Vending-Bench Arena | $5,838 | #13 of 32 | Official leaderboard | |
| Legal Research Bench | 38.5% | #36 of 88 | Official leaderboard | |
| MedScribe | 83% | #46 of 98 | Official leaderboard | |
| E-Commerce Bench | 2.59 | #9 of 19 | Official leaderboard | |
| Vals Finance Agent v2 | 51.5% | #40 of 80 | Official leaderboard | |
| JobBench | 44.5% | #8 of 16 | Official leaderboard | |
| APEX-Agents | 49.2% | #30 of 56 | Epoch AI | |
| Tax Agent Bench | 23.7% | #44 of 79 | Official leaderboard | |
| PolicyBench | 80.5% | #33 of 50 | Official leaderboard | |
| prinzbench | 25.0 | #19 of 26 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Vision Arena | 1,299 | #4 of 136 | Official leaderboard | |
| CharXiv | 91% | #5 of 67 | Model card | |
| ZeroBench | 15% | #20 of 66 | Official leaderboard | |
| Drone-Bench | 68.5% | #9 of 19 | Official leaderboard | |
| ParseBench | 63.3% | #22 of 42 | Official leaderboard | |
| GDP.pdf | 21% | #22 of 39 | Epoch AI | |
| ClockBench | 15% | #23 of 37 | Official leaderboard | |
| Furniture Assembly Benchmark | 33.3% | #24 of 35 | Epoch AI | |
| Blueprint-Bench 2 | 24.5% | #26 of 35 | Official leaderboard | |
| VoxelBench | 1,383 | #46 of 56 | Official leaderboard |
Safety and security
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SAGE (Vals) | 56.1% | #1 of 81 | Official leaderboard | |
| ACCoRD | 79.8 | #13 of 150 | Official leaderboard | |
| Conceptual Reasoning Index | 65.5 | #19 of 149 | Official leaderboard | |
| CAIS Risk Index | 37.8 | #10 of 65 | Official leaderboard | |
| LMCA | 52.2% | #22 of 139 | Epoch AI | |
| Vals Multimodal Index | 67.4% | #10 of 33 | Official leaderboard | |
| CAIS Text Capabilities Index | 38.0 | #23 of 66 | Official leaderboard | |
| CAIS Vision Capabilities Index | 51.2 | #27 of 53 | Official leaderboard | |
| Anthropic ECI (AECI) | 155.1 | #6 of 6 | Model card |
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. 74 of them aren't in the launch post.
System Card: Claude Opus 4.7 16 Apr 2026
Capabilities
- SWE-bench Verified 87.6
- SWE-Bench Pro 64.3
- SWE-bench Multilingual 80.5
- SWE-bench Multimodal 34.5
- Terminal-Bench 2.0 69.4
- BrowseComp 79.3
- MMMLU 91.5
- Humanity's Last Exam 46.9
- CharXiv 82.1
- OSWorld-Verified 78
- GPQA Diamond 94.2
- ScreenSpot-Pro 79.5
- OfficeQA 86.3
- OfficeQA Pro 80.6
- Vals Finance Agent 64.4
- MCP Atlas 77.3
- ARC-AGI 92
- ARC-AGI-2 75.8
- USAMO 2026 69.3
- GraphWalks
- OpenAI MRCR v2
- DeepSearchQA 89.1
- DRACO 77.7
- LAB-Bench FigQA 78.6
- Vending-Bench 2 10937
- GDPval-AA 1753
- Global MMLU 89.9
- MILU 89.9
- INCLUDE 87
- BioPipelineBench Verified (Anthropic) 83.6
- BioMysteryBench
- Structural Biology (Anthropic)
- Organic Chemistry (Anthropic) 77.2
- Phylogenetics (Anthropic) 79.6
- Protocols (Anthropic life sciences) 51.8
Hallucination
Safety
- Single-turn harmful request evaluations (Anthropic) 98
- Single-turn benign request evaluations (Anthropic) 0.28
- Higher-difficulty harmful request evaluations (Anthropic) 99
- Higher-difficulty benign request evaluations (Anthropic) 0.01
- Multi-turn testing (Anthropic)
- Child safety evaluations (Anthropic)
- Suicide and self-harm evaluations (Anthropic)
- Disordered eating evaluations (Anthropic)
- Election integrity evaluations (Anthropic)
- Malicious use of Claude Code (Anthropic) 91.2
- Malicious computer use (Anthropic) 89.3
Robustness
Alignment
- Impossible tasks (Anthropic) 45
- Agentic code behavior scores (Anthropic)
- Over-eager behavior on GUI computer use tasks (Anthropic) 21.1
- Anthropic automated behavioral audit
- Petri
- UK AISI unprompted sabotage evaluation 0
- UK AISI sabotage continuation evaluation 0
- Destructive production evaluation (Anthropic) 2.5
- Destructive actions (Anthropic)
- Adherence to its constitution (Anthropic) 0.74
- MASK 91.7
- Refusal to assist with AI safety R&D (Anthropic) 33.1
- Self-preference bias (Anthropic) 0.4
- DTBench