Overall score
69.7
Rank #5 · A confidence
Anthropic · stable
Anthropic's documented model for complex agentic coding and enterprise work.
Overall score
69.7
Rank #5 · A confidence
Capability
66.4
30 eligible benchmark families
Coverage
94%
Confidence A
Benchmarks scored
35
42 raw result rows
Model specification
Identity, configuration and price are preserved as dated source records.
Category profile
A category remains Not evaluated until an eligible source-qualified result is attached.
Benchmark scores
Best published score per benchmark. Open any row for the full per-benchmark leaderboard.
| Benchmark | Category | Score | Date |
|---|---|---|---|
| USAMO 2026Ranking weighted | mathematics | 96.7% | 2026-07-15 |
| τ²-Bench TelecomRanking weighted | agents | 94.4444% | 2026-07-15 |
| τ-benchRanking weighted | agents | 94.4% | 2026-07-15 |
| GPQA DiamondRanking weighted | reasoning | 93.6% | 2026-07-15 |
| CharXivRanking weighted | multimodal | 89.9% | 2026-07-15 |
| SWE-bench VerifiedRanking weighted | coding | 88.6% | 2026-07-15 |
| BenchLM Knowledge priorRanking weighted | research | 88.1% | 2026-07-15 |
| Terminal-BenchRanking weighted | agents | 84.64% | 2026-07-15 |
| BrowseCompRanking weighted | agents | 84.3% | 2026-07-15 |
| OSWorld-VerifiedRanking weighted | agents | 83.4% | 2026-07-15 |
| MCP AtlasRanking weighted | agents | 82.2% | 2026-07-15 |
| LiveBenchRanking weighted | reasoning | 77.22% | 2026-07-15 |
| BenchLM Math priorRanking weighted | mathematics | 73.6% | 2026-07-15 |
| BenchLM Coding priorRanking weighted | coding | 73.49% | 2026-07-15 |
| SWE-Bench ProRanking weighted | coding | 69.2% | 2026-07-15 |
| BenchLM Agentic priorRanking weighted | agents | 68.06% | 2026-07-15 |
| AA Long Context ReasoningRanking weighted | research | 67.6667% | 2026-07-15 |
| BenchLM Multimodal priorRanking weighted | multimodal | 66.2% | 2026-07-15 |
| CursorBenchRanking weighted | coding | 62.3% | 2026-07-15 |
| IFBench | instruction following | 62.2% | 2026-07-15 |
| IFEval | instruction following | 62.2% | 2026-07-15 |
| ToolathlonRanking weighted | agents | 59.9% | 2026-07-15 |
| Terminal-Bench HardRanking weighted | agents | 58.3333% | 2026-07-15 |
| Long-Horizon-Terminal-BenchRanking weighted | agents | 58.3% | 2026-07-15 |
| Humanity's Last ExamRanking weighted | reasoning | 57.9% | 2026-07-15 |
| GDPval-AA v2Ranking weighted | agents | 55.006% | 2026-07-15 |
| SciCodeRanking weighted | coding | 53.5% | 2026-07-15 |
| AA AutomationBenchRanking weighted | agents | 48.5% | 2026-07-15 |
| FrontierMathRanking weighted | mathematics | 47.241% | 2026-07-15 |
| EnterpriseOps-Gym-AARanking weighted | agents | 43.957% | 2026-07-15 |
| τ³-BankingRanking weighted | agents | 27.6289% | 2026-07-15 |
| AA-OmniscienceRanking weighted | research | 27.4333 index | 2026-07-15 |
| CritPtRanking weighted | reasoning | 20.8571% | 2026-07-15 |
| OSWorldRanking weighted | agents | 20.6% | 2026-07-15 |
| Harvey LAB-AARanking weighted | agents | 7.5% | 2026-07-15 |
Showing the best of 42 raw result rows across 35 benchmark families.
Price and efficiency
Operational values are shown only when the source publishes a comparable measurement or a valid derivation.
Verification summary
News and X posts are structurally excluded from this model's evidence graph.
Capability ranking is complete for the current snapshot. The last source check for this exact model was 2026-07-14.
Related models
Exact variants remain separate even when they share a provider or family.
Claude 5 · stable
Claude Sonnet 5Claude Sonnet 5 · stable
Claude Haiku 4.5Claude Haiku 4 · stable