Anthropic
availableShows if the model has enough results for an index.Claude 4 Sonnet
Claude 4 Sonnet is a non-reasoning model from Anthropic. 13 benchmarks count toward its score, in 6 categories.
IndexOverall score. 50 is the middle.42.5 ±6.0
CoverageShare of the index weight with results.85%
SpeedOutput tokens per second.40/s
Input / 1MUS dollars per 1M input tokens.$3
Output / 1MUS dollars per 1M output tokens.$15
ContextMaximum tokens in one request.200K
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
13 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. |
|---|---|---|---|---|---|---|
| Software Engineering Benchmark Verified | Coding | 72.7% | 55.8 | — | — | Carlos E. Jimenez et al. |
| SWE-bench Verified | Coding | 70.8% | 54.3 | Moatless Tools | 1 Sept 2026 | SWE-bench team |
| SWE-bench Verified | Coding | 68.5% | 52.5 | OpenHands | 26 Feb 2026 | SWE-bench team |
| Artificial Analysis GPQA Diamond | Knowledge | 68.3% | 38.6 | — | — | Artificial Analysis |
| Artificial Analysis MMMU-Pro | Multimodal | 62.4% | 43.4 | — | — | Artificial Analysis |
| SWE-bench Lite | Coding | 57.5% | 52.4 | SWE-agent | 11 Sept 2025 | SWE-bench team |
| SWE-bench Lite | Coding | 56.7% | 51.7 | SWE-agent | 11 Sept 2025 | SWE-bench team |
| τ²-Bench Tool-Agent-User Evaluation | Agentic | 52.3% | 36.4 | — | — | Victor Barres et al. |
| Artificial Analysis IFBench | Instruction | 45.4% | 35.2 | — | — | Artificial Analysis |
| Artificial Analysis Long Context Reasoning | Reasoning | 44.0% | 38.7 | — | — | Artificial Analysis |
| Gert Labs Composite Game Benchmark | Agentic | 39.7% | 47.8 | — | — | Gert Labs |
| SWE-bench Multimodal | Multimodal | 35.0% | — | Refact.ai Agent | 17 Nov 2025 | SWE-bench team |
| Artificial Analysis Omniscience Accuracy | Knowledge | 22.7% | 42.0 | — | — | Artificial Analysis |
| JobBench | Agentic | 18.4% | 46.8 | — | — | Yuetai Li et al. |
| Artificial Analysis Intelligence Index | Knowledge | 16.6% | 43.2 | — | — | Artificial Analysis |
| Artificial Analysis Humanity's Last Exam | Knowledge | 4.3% | 29.2 | — | — | Artificial Analysis |
| Critical Physics Tasks | Reasoning | 1.1% | 40.7 | — | — | Artificial Analysis |
13 benchmarks count, from 16 of 17 results. A grey row does not count. Too few models took that benchmark.