0 vs 2 benchmarks won
Anthropic Claude 3.7 Sonnet | Moonshot AI Kimi K2.5 | |
|---|---|---|
| Overview | ||
| Company | Anthropic | Moonshot AI |
| Release date | Feb 24 2025 | Jan 27 2026 |
| Model type | — | — |
| Open source | No | Yes |
| Specifications | ||
Parameters | — | 1T |
Context window | — | 256k |
| Benchmarks | ||
Science reasoning GPQA Diamond | 68% | 87.6% |
Software engineering SWE-Bench Verified | 62.3% | 76.8% |
Multimodal understanding MMMU | — | — |
| Timeline | ||
| Release gap | Claude 3.7 Sonnet shipped 337 days before Kimi K2.5 | |
Kimi K2.5 leads Claude 3.7 Sonnet on 2 of the tracked benchmarks (GPQA Diamond, SWE-Bench Verified, MMMU). Claude 3.7 Sonnet shipped 337 days before Kimi K2.5, so benchmark comparisons should account for the intervening progress.
Kimi K2.5 is an open-source / open-weight model; Claude 3.7 Sonnet is proprietary.
On GPQA Diamond, Kimi K2.5 scores 87.6%, 19.6 points above Claude 3.7 Sonnet at 68%. On SWE-Bench Verified, Kimi K2.5 scores 76.8%, 14.5 points above Claude 3.7 Sonnet at 62.3%.