0 vs 2 benchmarks won
Anthropic Claude 3.5 Haiku | Anthropic Claude Sonnet 4 | |
|---|---|---|
| Overview | ||
| Company | Anthropic | Anthropic |
| Release date | Oct 22 2024 | May 22 2025 |
| Model type | — | — |
| Open source | No | No |
| Specifications | ||
Parameters | — | — |
Context window | — | — |
| Benchmarks | ||
Science reasoning GPQA Diamond | 41.6% | 75.4% |
Software engineering SWE-Bench Verified | 40.6% | 72.7% |
Multimodal understanding MMMU | — | — |
| Timeline | ||
| Release gap | Claude 3.5 Haiku shipped 212 days before Claude Sonnet 4 | |
Claude Sonnet 4 leads Claude 3.5 Haiku on 2 of the tracked benchmarks (GPQA Diamond, SWE-Bench Verified, MMMU). Claude 3.5 Haiku shipped 212 days before Claude Sonnet 4, 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 scores 75.4%, 33.8 points above Claude 3.5 Haiku at 41.6%. On SWE-Bench Verified, Claude Sonnet 4 scores 72.7%, 32.1 points above Claude 3.5 Haiku at 40.6%.