2 vs 0 benchmarks won
Anthropic Claude Opus 4 | Moonshot AI Kimi K2 | |
|---|---|---|
| Overview | ||
| Company | Anthropic | Moonshot AI |
| Release date | May 22 2025 | Jul 11 2025 |
| Model type | — | — |
| Open source | No | Yes |
| Specifications | ||
Parameters | — | 1T |
Context window | — | 128k |
| Benchmarks | ||
Science reasoning GPQA Diamond | 79.6% | 75.1% |
Software engineering SWE-Bench Verified | 72.5% | 65.8% |
Multimodal understanding MMMU | — | — |
| Timeline | ||
| Release gap | Claude Opus 4 shipped 50 days before Kimi K2 | |
Claude Opus 4 leads Kimi K2 on 2 of the tracked benchmarks (GPQA Diamond, SWE-Bench Verified, MMMU). Claude Opus 4 shipped 50 days before Kimi K2, so benchmark comparisons should account for the intervening progress.
Kimi K2 is an open-source / open-weight model; Claude Opus 4 is proprietary.
On GPQA Diamond, Claude Opus 4 scores 79.6%, 4.5 points above Kimi K2 at 75.1%. On SWE-Bench Verified, Claude Opus 4 scores 72.5%, 6.7 points above Kimi K2 at 65.8%.