Anthropic
availableShows if the model has enough results for an index.Claude Opus 4.7
Claude Opus 4.7 is a non-reasoning model from Anthropic. 41 benchmarks count toward its score, in 7 categories.
IndexOverall score. 50 is the middle.63.3 ±3.0
CoverageShare of the index weight with results.95%
SpeedOutput tokens per second.43/s
Input / 1MUS dollars per 1M input tokens.$5 batch $2.5
Output / 1MUS dollars per 1M output tokens.$25 batch $12.5 US dollars per 1M output tokens in a batch.
ContextMaximum tokens in one request.1M
EloLMArena rating and rank.1483 (#12)
50 is the middle of the board. The range shows the doubt in the index. Batch work costs less.
61,128 votes. Elo shows what people prefer. It does not change the score.
CapabilitiesScore per category. 50 is the middle.
50 is the middleResults
41 counted| BenchmarkThe test name. | CategoryThe capability that the test measures. | ResultThe score from the publisher. | IndexThis result on the index scale. | RunThe settings of the run. | DateDate of the result. | Published byThe source of the result. |
|---|---|---|---|---|---|---|
| AIME | Math | 96.3% | 58.6 | — | 16 Apr 2026 | Vals AI |
| LiveBench Mathematics | Math | 92.9% | 72.0 | xhigh effort | 25 Jun 2026 | LiveBench |
| GPQA Diamond | Knowledge | 90.2% | 61.5 | — | 1 Sept 2026 | Vals AI |
| MMLU Pro | Knowledge | 89.9% | 62.2 | — | 1 Sept 2026 | Vals AI |
| Artificial Analysis GPQA Diamond | Knowledge | 88.5% | 59.2 | — | — | Artificial Analysis |
| LiveBench Reasoning | Reasoning | 87.2% | 77.1 | xhigh effort | 25 Jun 2026 | LiveBench |
| OTIS Mock AIME 2024-2025 | Math | 86.7% | 60.0 | max effort | — | Epoch AI |
| GPQA diamond | Knowledge | 86.4% | 58.0 | max effort | — | Epoch AI |
| MMMU Pro | Multimodal | 85.5% | 66.6 | — | 1 Sept 2026 | Vals AI |
| LiveCodeBench | Coding | 85.1% | 61.7 | — | 1 Sept 2026 | Vals AI |
| SWE-Bench verified | Coding | 83.5% | 64.5 | max effort | — | Epoch AI |
| React Native Evals | Coding | 82.8% | 63.3 | — | — | Callstack |
| LiveBench Coding | Coding | 82.1% | 73.9 | xhigh effort | 25 Jun 2026 | LiveBench |
| SWE-bench | Coding | 82.0% | 63.3 | — | 1 Sept 2026 | Vals AI |
| LiveBench Data Analysis | Reasoning | 78.3% | 64.7 | xhigh effort | 25 Jun 2026 | LiveBench |
| LiveBench Language | Knowledge | 77.9% | 64.6 | xhigh effort | 25 Jun 2026 | LiveBench |
| Artificial Analysis MMMU-Pro | Multimodal | 76.4% | 60.5 | — | — | Artificial Analysis |
| Artificial Analysis Long Context Reasoning | Reasoning | 75.7% | 60.6 | — | — | Artificial Analysis |
| τ²-Bench Tool-Agent-User Evaluation | Agentic | 74.0% | 51.9 | — | — | Victor Barres et al. |
| Vibe Code Bench v1.1 | Coding | 71.0% | 71.7 | OpenHands | 21 Sept 2026 | Vals AI |
| FrontierMath-Tiers-1-3-v2-Private | Math | 70.2% | 70.9 | max effort | — | Epoch AI |
| Terminal-Bench 2.0 | Agentic | 68.5% | 71.6 | — | 4 Jun 2026 | Vals AI |
| Terminal-Bench 2.1 | Agentic | 68.5% | 64.5 | — | 21 Sept 2026 | Vals AI |
| LiveBench Instruction Following | Instruction | 66.7% | 65.4 | xhigh effort | 25 Jun 2026 | LiveBench |
| Gert Labs Composite Game Benchmark | Agentic | 65.6% | 70.6 | — | — | Gert Labs |
| SimpleQA Verified | Knowledge | 51.7% | 69.4 | xhigh effort | — | Epoch AI |
| LiveBench Agentic Coding | Agentic | 50.7% | 64.4 | xhigh effort | 25 Jun 2026 | LiveBench |
| IOI v1 | Coding | 47.1% | 66.7 | — | 9 Aug 2026 | Vals AI |
| Artificial Analysis Omniscience Accuracy | Knowledge | 44.7% | 69.2 | — | — | Artificial Analysis |
| Code Migration | Coding | 43.9% | 73.0 | — | 21 Sept 2026 | Vals AI |
| FrontierMath-2025-02-28-Private | Math | 43.8% | 71.5 | xhigh effort | — | Epoch AI |
| Artificial Analysis IFBench | Instruction | 43.6% | 33.4 | — | — | Artificial Analysis |
| τ²-bench Banking | Agentic | 40.2% | 27.7 | max effort · Sierra | 4 Aug 2026 | Sierra Research |
| FrontierCode 1.1 Main | Coding | 38.5% | 67.7 | — | — | Cognition |
| Furniture Assembly | Reasoning | 33.3% | 63.2 | max effort | — | Epoch AI |
| Artificial Analysis Humanity's Last Exam | Knowledge | 33.3% | 60.7 | — | — | Artificial Analysis |
| FrontierMath-Tier-4-v2-Private | Math | 31.7% | 63.2 | max effort | — | Epoch AI |
| MirrorCode | Coding | 31.1% | — | high effort | — | Epoch AI |
| Artificial Analysis Intelligence Index | Knowledge | 30.9% | 61.1 | — | — | Artificial Analysis |
| Mystery Game Puzzles | Reasoning | 28.0% | 63.7 | max effort | — | Epoch AI |
| FrontierMath-Tier-4-2025-07-01-Private | Math | 22.9% | 68.3 | xhigh effort | — | Epoch AI |
| ResearchClawBench | Agentic | 20.7% | — | — | — | InternScience |
| EBR-bench | Reasoning | 19.0% | 62.4 | max effort | — | Epoch AI |
| OSWorld 2.0 | Agentic | 13.9% | 62.6 | — | — | Mengqi Yuan et al. |
| Chess Puzzles | Reasoning | 7.0% | 32.6 | max effort | — | Epoch AI |
| ApprenticeBench: end-to-end computer use, continual learning, and long-horizon agency on a real accounts-payable job | Agentic | 7.0% | 64.0 | — | — | NeoCognition |
| Critical Physics Tasks | Reasoning | 5.1% | 49.1 | — | — | Artificial Analysis |
| ProgramBench | Coding | 0.0% | — | — | 21 Sept 2026 | Vals AI |
41 benchmarks count, from 45 of 48 results. A grey row does not count. Too few models took that benchmark.