0 vs 2 benchmarks won
Anthropic Claude Sonnet 4 | Anthropic Claude Sonnet 4.6 | |
|---|---|---|
| Overview | ||
| Company | Anthropic | Anthropic |
| Release date | May 22 2025 | Feb 17 2026 |
| Model type | — | — |
| Open source | No | No |
| Specifications | ||
Parameters | — | — |
Context window | — | — |
| Benchmarks | ||
Science reasoning GPQA Diamond | 75.4% | 89.9% |
Software engineering SWE-Bench Verified | 72.7% | 79.6% |
Multimodal understanding MMMU | — | — |
| Timeline | ||
| Release gap | Claude Sonnet 4 shipped 271 days before Claude Sonnet 4.6 | |
Claude Sonnet 4.6 leads Claude Sonnet 4 on 2 of the tracked benchmarks (GPQA Diamond, SWE-Bench Verified, MMMU). Claude Sonnet 4 shipped 271 days before Claude Sonnet 4.6, so benchmark comparisons should account for the intervening progress.
Published specifications for these two models are limited — see each model page for the latest details.
On GPQA Diamond, Claude Sonnet 4.6 scores 89.9%, 14.5 points above Claude Sonnet 4 at 75.4%. On SWE-Bench Verified, Claude Sonnet 4.6 scores 79.6%, 6.9 points above Claude Sonnet 4 at 72.7%.