Claude Haiku 4.5 ranks first on 4 of the 137 benchmarks it has scores on.
Closed weights Small
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 | |
|---|---|---|---|---|
| Phare | 83.2% | #1 of 62 | Official leaderboard | |
| BullshitBench | 87.3% | #5 of 107 | Official leaderboard | |
| BullshitBench V2 | 77% | #15 of 122 | Official leaderboard | |
| WritingBench | 77.1 | #14 of 45 | Official leaderboard | |
| LMArena (Elo) | 1,415 | #112 of 333 | Official leaderboard | |
| Judgemark v2.1 | 79.7 | #16 of 42 | Official leaderboard | |
| Longform Creative Writing (EQ-Bench) | 65.0 | #55 of 119 | Official leaderboard | |
| Vectara Hallucination Leaderboard (HHEM) | 4.50% | #74 of 148 | Official leaderboard | |
| HELM Arabic | 78.8 | #14 of 28 | Official leaderboard | |
| Vectara Hallucination Leaderboard (2025 dataset) | 9.80% | #58 of 104 | Official leaderboard | |
| MMMLU | 83% | #35 of 62 | Model card | |
| HELM Arabic Enterprise | 63.6 | #14 of 23 | Official leaderboard | |
| SEA-HELM | 64.9 | #39 of 62 | Official leaderboard | |
| MultiChallenge | 50.5% | #29 of 35 | Official leaderboard | |
| Extended NYT Connections | 14.3% | #88 of 97 | Official leaderboard | |
| SimpleQA Verified | 13.2% | #77 of 84 | Epoch AI | |
| EQ-Bench 4 | 1,053 | #30 of 32 | Official leaderboard | |
| Document Arena | 1,419 | #35 of 37 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MATH Level 5 | 96.4% | #9 of 94 | Epoch AI | |
| AIME 2025 | 96.3% | #18 of 143 | Model card | |
| GPQA Diamond | 71.2% | #107 of 219 | Epoch AI | |
| OTIS Mock AIME 2024-2025 | 66.7% | #107 of 202 | Epoch AI | |
| ForecastBench | 63.2 | #49 of 86 | Official leaderboard | |
| DTBench | 56% | #108 of 178 | Official leaderboard | |
| FrontierMath | 5.90% | #43 of 70 | Epoch AI | |
| ARC-AGI | 47.7% | #66 of 92 | Official leaderboard | |
| ARC-AGI-2 | 4.03% | #70 of 95 | Official leaderboard | |
| FrontierMath Tier 4 | 2.08% | #43 of 55 | Epoch AI | |
| CritPt | 0% | #143 of 160 | Epoch AI | |
| CORE-Bench | 11.1% | #16 of 17 | Official leaderboard | |
| BixBench3 | 1.50 | #19 of 20 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SWE-bench Verified | 66.6% | #18 of 60 | Official leaderboard | |
| ProgramBench | 0% | #12 of 25 | Official leaderboard | |
| SWE-Bench Pro | 39.5% | #12 of 24 | Official leaderboard | |
| WeirdML | 45.4% | #62 of 112 | Official leaderboard | |
| Vibe Code Bench v1 | 11.7% | #11 of 17 | Official leaderboard | |
| ALE-Bench | 653 | #88 of 121 | Epoch AI | |
| IOI (Vals) | 6.17% | #40 of 55 | Official leaderboard | |
| WebDev Arena | 1,329 | #88 of 105 | Epoch AI | |
| Code Arena | 1,330 | #104 of 124 | Official leaderboard | |
| Vibe Code Bench | 11.4% | #97 of 115 | Official leaderboard | |
| Code Migration | 7.55% | #66 of 78 | Official leaderboard | |
| SWE-bench Multilingual | 64.7% | #11 of 13 | Official leaderboard | |
| OTelBench | 5.80% | #12 of 14 | Official leaderboard | |
| SWE-bench Verified (hard subset) | 36.6% | #8 of 9 | Model card | |
| SWE-Bench Pro V2 Hard | 25.5% | #16 of 16 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| BFCL V4 | 68.7% | #6 of 65 | Official leaderboard | |
| Terminal-Bench | 35.5% | #5 of 18 | Epoch AI | |
| CompileBench | 86.7% | #7 of 20 | Official leaderboard | |
| PinchBench | 71.6% | #17 of 48 | Official leaderboard | |
| SnakeBench | 26 | #69 of 183 | Official leaderboard | |
| MCP-Universe | 26.4% | #13 of 30 | Official leaderboard | |
| BALROG | 31.2% | #16 of 35 | Official leaderboard | |
| LLM Chess | 127 | #58 of 124 | Official leaderboard | |
| ClawBench | 30.1% | #3 of 6 | Official leaderboard | |
| CEO-Bench | $59,625 | #10 of 19 | Official leaderboard | |
| PACT | 1,510 | #15 of 26 | Official leaderboard | |
| tau2-bench | 76.6% | #42 of 69 | Model card | |
| PinchBench 2.0 | 67.7% | #35 of 57 | Official leaderboard | |
| DeepResearch Bench | 45.5% | #15 of 24 | Epoch AI | |
| Terminal-Bench 2.0 | 35.5% | #28 of 43 | Official leaderboard | |
| Toolathlon | 26.2% | #36 of 50 | Official leaderboard | |
| ITBench | 42.1% | #10 of 13 | Official leaderboard | |
| GAIA | 23.6% | #44 of 57 | Official leaderboard | |
| LisanBench | 137.0 | #103 of 133 | Official leaderboard | |
| BalatroBench | 5.07 | #14 of 16 | Official leaderboard | |
| Game Arena Chess Openings | 102 | #10 of 11 | Official leaderboard | |
| Chess Puzzles | 8.00% | #76 of 83 | Epoch AI | |
| OSWorld-Verified | 50.7% | #34 of 37 | Model card | |
| SkillsBench | 30.1% | #20 of 21 | Official leaderboard | |
| MCP Atlas | 40.2% | #34 of 34 | Official leaderboard | |
| DeepSearchQA | 12.8% | #10 of 10 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MedScribe | 85.2% | #30 of 98 | Official leaderboard | |
| Vals Finance Agent v1 | 43% | #16 of 47 | Official leaderboard | |
| CorpFin v2 | 60.6% | #66 of 119 | Official leaderboard | |
| CaseLaw v2 | 56.5% | #30 of 53 | Official leaderboard | |
| LegalBench | 81.2% | #79 of 138 | Official leaderboard | |
| MortgageTax | 62.2% | #53 of 86 | Official leaderboard | |
| Vals Finance Agent | 46.9% | #32 of 49 | Official leaderboard | |
| FailureSensorIQ | 61.8% | #16 of 24 | Official leaderboard | |
| Harvey LAB (Legal Agent Benchmark) | 0.83% | #59 of 80 | Official leaderboard | |
| TaxEval v2 | 67.5% | #97 of 129 | Official leaderboard | |
| EnterpriseClawBench | 51.3% | #7 of 9 | Official leaderboard | |
| Vending-Bench Arena | $1,701 | #26 of 32 | Official leaderboard | |
| Public Benefits Bench | 54.3% | #42 of 50 | Official leaderboard | |
| MedCode | 32.7% | #81 of 96 | Official leaderboard | |
| Vending-Bench 2 | $459 | #57 of 66 | Official leaderboard | |
| Legal Research Bench | 10.6% | #77 of 88 | Official leaderboard | |
| Vals Finance Agent v2 | 31% | #71 of 80 | Official leaderboard | |
| ProfBench | 39.6% | #65 of 72 | Official leaderboard | |
| PolicyBench | 74.5% | #46 of 50 | Official leaderboard | |
| Excel Modeling Benchmark | 22.7% | #70 of 76 | Official leaderboard | |
| Tax Agent Bench | 5.86% | #74 of 79 | Official leaderboard | |
| EEBench | 3.50% | #26 of 26 | Official leaderboard | |
| RCM-Bench-1 | 70.2% | #20 of 20 | Official leaderboard | |
| TaxCalcBench | 13.7% | #17 of 17 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MADQA | 68.2% | #6 of 16 | Official leaderboard | |
| MMMU | 73.2% | #47 of 97 | Model card | |
| ParseBench | 53.1% | #31 of 42 | Official leaderboard | |
| Blueprint-Bench 2 | 0% | #31 of 35 | Official leaderboard |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| AIR-Bench 2024 | 93.2% | #1 of 82 | Official leaderboard | |
| KORA | 70.2% | #2 of 42 | Official leaderboard | |
| Impossible tasks (Anthropic) | 30% | #1 of 11 | Model card | |
| Single-turn harmful request evaluations (Anthropic) | 99.4% | #3 of 18 | Model card | |
| Agentic coding malicious requests (Anthropic) | 100% | #1 of 5 | Model card | |
| Single-turn benign request evaluations (Anthropic) | 0.02% | #4 of 18 | Model card | |
| HELM Safety | 95.5 | #21 of 75 | Official leaderboard | |
| MCP prompt injection evaluation (Anthropic) | 92.5% | #1 of 2 | Model card | |
| SimpleSafetyTests | 98.8% | #41 of 75 | Official leaderboard | |
| Computer use prompt injection evaluation (Anthropic) | 72.2% | #3 of 5 | Model card | |
| LAB-Bench CloningScenarios | 45% | #7 of 11 | Model card | |
| Cybench | 46.9% | #11 of 17 | Model card | |
| BrowseSafe++ | 72.8% | #11 of 15 | Official leaderboard | |
| Reward-hack-prone coding tasks (Anthropic) | 6.00% | #6 of 8 | Model card | |
| Human Pathogen Capabilities Test v2 (HPCT v2) | 42.5% | #55 of 71 | Official leaderboard | |
| Anthropic Red Team | 96.9% | #60 of 75 | Official leaderboard | |
| LAB-Bench FigQA | 49% | #8 of 10 | Model card | |
| Average in-eval reward hacking rate (Anthropic) | 15.5% | #4 of 5 | Model card | |
| LAB-Bench SeqQA | 66% | #10 of 12 | Model card | |
| BinaryAudit | 10.5% | #17 of 20 | Official leaderboard | |
| LAB-Bench ProtocolQA | 69% | #13 of 15 | Model card | |
| BBQ (Bias Benchmark for QA) | 71.2% | #21 of 24 | Model card | |
| ExploitBench | 14% | #8 of 9 | Official leaderboard | |
| Virology Capabilities Test (VCT) | 32% | #26 of 28 | Model card | |
| Gray Swan Agent Red Teaming (ART) benchmark | 51.1% | #13 of 14 | Model card | |
| DeepsecBench | 4.69 | #27 of 28 | Official leaderboard | |
| PCT (Polarized Contrastive Pairs) | 62.2% | #45 of 46 | Official leaderboard | |
| Malicious use of Claude Code (Anthropic) | 69.4% | #16 of 16 | Model card | |
| CRSBench | 31.4% | #5 of 5 | Official leaderboard | |
| Tool use prompt injection evaluation (Anthropic) | 93.4% | #2 of 2 | Model card |
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| HELM Capabilities | 71.7 | #20 of 60 | Official leaderboard | |
| ACCoRD | 66.7 | #60 of 150 | Official leaderboard | |
| LMCA | 30.9% | #79 of 139 | Epoch AI | |
| Conceptual Reasoning Index | 43.1 | #89 of 149 | Official leaderboard | |
| CAIS Risk Index | 54.3 | #39 of 65 | Official leaderboard | |
| SAGE (Vals) | 31.8% | #71 of 81 | Official leaderboard | |
| CAIS Vision Capabilities Index | 37.5 | #48 of 53 | Official leaderboard | |
| CAIS Text Capabilities Index | 12.4 | #60 of 66 | Official leaderboard | |
| Vals Multimodal Index | 42.9% | #30 of 33 | 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. 25 of them aren't in the launch post.