Anthropic
availableShows if the model has enough results for an index.Claude Opus 4.7 (Adaptive)
Claude Opus 4.7 (Adaptive) is a reasoning model from Anthropic in the Claude Opus 4.7 family. 24 benchmarks count toward its score, in 6 categories.
IndexOverall score. 50 is the middle.66.2 ±4.8
CoverageShare of the index weight with results.85%
SpeedOutput tokens per second.—
Input / 1MUS dollars per 1M input tokens.$5
Output / 1MUS dollars per 1M output tokens.$25
ContextMaximum tokens in one request.1M
EloLMArena rating and rank.N/A
50 is the middle of the board. The range shows the doubt in the index.
CapabilitiesScore per category. 50 is the middle.
50 is the middleResults
24 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. |
|---|---|---|---|---|---|---|
| Graduate-Level Google-Proof Q&A | Knowledge | 94.2% | 65.2 | — | — | David Rein et al. |
| GPQA Diamond | Knowledge | 94.2% | 65.2 | — | — | David Rein et al. |
| Artificial Analysis GPQA Diamond | Knowledge | 91.4% | 62.2 | — | — | Artificial Analysis |
| CharXiv Reasoning | Multimodal | 91.0% | 69.7 | — | — | CharXiv authors |
| τ²-Bench Tool-Agent-User Evaluation | Agentic | 88.6% | 62.4 | — | — | Victor Barres et al. |
| Software Engineering Benchmark Verified | Coding | 87.6% | 67.8 | — | — | Carlos E. Jimenez et al. |
| CharXiv Reasoning without tools | Multimodal | 82.1% | — | — | — | CharXiv authors |
| BrowseComp | Agentic | 79.3% | 66.4 | — | — | OpenAI |
| Artificial Analysis MMMU-Pro | Multimodal | 78.8% | 63.4 | — | — | Artificial Analysis |
| Artificial Analysis Long Context Reasoning | Reasoning | 78.7% | 62.7 | — | — | Artificial Analysis |
| OSWorld-Verified | Agentic | 78.0% | 66.4 | — | — | Tianbao Xie et al. |
| MCP Atlas | Agentic | 77.3% | 66.4 | — | — | OpenAI |
| Artificial Analysis Coding Index | Coding | 73.6% | 70.8 | — | — | Artificial Analysis |
| CyberGym | Agentic | 73.1% | 66.2 | — | — | Zhun Wang et al. |
| SWE-bench Pro | Coding | 64.3% | 66.2 | — | — | Xiang Deng et al. |
| OpenAI MRCR v2 8-needle 128K-256K | Reasoning | 59.2% | — | — | — | OpenAI |
| Artificial Analysis IFBench | Instruction | 58.6% | 48.6 | — | — | Artificial Analysis |
| Humanity's Last Exam | Knowledge | 54.7% | 75.1 | — | — | Center for AI Safety et al. |
| Artificial Analysis Omniscience Accuracy | Knowledge | 48.9% | 74.4 | — | — | Artificial Analysis |
| Humanity's Last Exam without tools | Knowledge | 46.9% | 68.5 | — | — | OpenAI |
| Artificial Analysis ITBench-AA | Agentic | 46.7% | — | — | — | Artificial Analysis |
| JobBench | Agentic | 45.9% | 65.7 | — | — | Yuetai Li et al. |
| OfficeQA Pro | Multimodal | 43.6% | 55.9 | — | — | OfficeQA Pro authors |
| Artificial Analysis Humanity's Last Exam | Knowledge | 42.3% | 70.4 | — | — | Artificial Analysis |
| GDPval-AA normalized | Agentic | 41.9% | 67.7 | — | — | Artificial Analysis |
| Artificial Analysis Intelligence Index | Knowledge | 40.7% | 73.3 | — | — | Artificial Analysis |
| Artificial Analysis Agentic Index | Agentic | 39.5% | 69.4 | — | — | Artificial Analysis |
| OSWorld 2.0 | Agentic | 18.2% | 64.6 | — | — | Mengqi Yuan et al. |
| Critical Physics Tasks | Reasoning | 12.0% | 63.5 | — | — | Artificial Analysis |
24 benchmarks count, from 26 of 29 results. A grey row does not count. Too few models took that benchmark.