Claude Sonnet 4
Claude Sonnet 4, part of the Claude 4 family, is a significant upgrade to Claude Sonnet 3.7. It excels in coding (72.7% on SWE-bench) and reasoning, responding more precisely to instructions. Sonnet 4 offers an optimal mix of capability and practicality, with enhanced steerability, and supports extended thinking with tool use.
Benchmark results
| Benchmark | Score | Tags | Source |
|---|---|---|---|
| AIME 2025 | 70.5% | self-reported llm-stats | link → |
| GPQA | 75.4% | self-reported llm-stats | link → |
| MMMLU | 86.5% | self-reported llm-stats | link → |
| MMMU | 74.4% | self-reported llm-stats | link → |
| SWE-Bench Verified | 72.7% | self-reported llm-stats | link → |
| TAU-bench Airline | 60.0% | self-reported llm-stats | link → |
| TAU-bench Retail | 80.5% | self-reported llm-stats | link → |
| Terminal-Bench | 35.5% | self-reported llm-stats | link → |