Overall score
58.2
Rank #30 · A confidence
Anthropic · stable
Anthropic's Claude Opus 4.5 model tracked via the BenchLM July 2026 leaderboard for capability comparison.
Overall score
58.2
Rank #30 · A confidence
Capability
69.8
29 eligible benchmark families
Coverage
78%
Confidence A
Benchmarks scored
35
36 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 |
|---|---|---|---|
| AIME 2026Ranking weighted | mathematics | 95.1% | 2026-07-15 |
| IFEval | instruction following | 90.9% | 2026-07-15 |
| MMLU-ProRanking weighted | research | 89.5% | 2026-07-15 |
| τ²-Bench TelecomRanking weighted | agents | 89.4737% | 2026-07-15 |
| GPQA DiamondRanking weighted | reasoning | 87% | 2026-07-15 |
| τ-benchRanking weighted | agents | 86.3% | 2026-07-15 |
| HMMT Feb 2026Ranking weighted | mathematics | 85.3% | 2026-07-15 |
| LiveCodeBenchRanking weighted | coding | 84.8% | 2026-07-15 |
| CybenchRanking weighted | agents | 82% | 2026-07-15 |
| BenchLM Knowledge priorRanking weighted | research | 80.9% | 2026-07-15 |
| SWE-bench VerifiedRanking weighted | coding | 80.9% | 2026-07-15 |
| LiveBenchRanking weighted | reasoning | 75.96% | 2026-07-15 |
| AA Long Context ReasoningRanking weighted | research | 74% | 2026-07-15 |
| MMMU-ProRanking weighted | multimodal | 70.6% | 2026-07-15 |
| SuperGPQARanking weighted | reasoning | 70.6% | 2026-07-15 |
| CharXivRanking weighted | multimodal | 68.5% | 2026-07-15 |
| BenchLM Reasoning priorRanking weighted | reasoning | 67.5% | 2026-07-15 |
| OSWorldRanking weighted | agents | 66.3% | 2026-07-15 |
| OSWorld-VerifiedRanking weighted | agents | 66.3% | 2026-07-15 |
| LongBench v2 | long context | 64.4% | 2026-07-15 |
| BenchLM Multimodal priorRanking weighted | multimodal | 59.3% | 2026-07-15 |
| BenchLM Coding priorRanking weighted | coding | 58.72% | 2026-07-15 |
| BenchLM Math priorRanking weighted | mathematics | 58.7% | 2026-07-15 |
| IFBench | instruction following | 58% | 2026-07-15 |
| SWE-Bench ProRanking weighted | coding | 57.1% | 2026-07-15 |
| BenchLM Agentic priorRanking weighted | agents | 55.56% | 2026-07-15 |
| SciCodeRanking weighted | coding | 47% | 2026-07-15 |
| Terminal-Bench HardRanking weighted | agents | 46.9697% | 2026-07-15 |
| ToolathlonRanking weighted | agents | 43.5% | 2026-07-15 |
| MCP AtlasRanking weighted | agents | 42.3% | 2026-07-15 |
| Long-Horizon-Terminal-BenchRanking weighted | agents | 40.9% | 2026-07-15 |
| Humanity's Last ExamRanking weighted | reasoning | 30.8% | 2026-07-15 |
| FrontierMathRanking weighted | mathematics | 20.69% | 2026-07-15 |
| AA-OmniscienceRanking weighted | research | 13.2667 index | 2026-07-15 |
| CritPtRanking weighted | reasoning | 4.5714% | 2026-07-15 |
Showing the best of 36 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.
Missing publication dimensions: efficiency, value. The last source check for this exact model was 2026-07-16.
Related models
Exact variants remain separate even when they share a provider or family.
Claude 5 · stable
Claude Opus 4.8Claude Opus 4 · stable
Claude Sonnet 5Claude Sonnet 5 · stable