2 vs 0 benchmarks won
Anthropic Claude Sonnet 4.5 | Moonshot AI Kimi K2 (0905) | |
|---|---|---|
| Overview | ||
| Company | Anthropic | Moonshot AI |
| Release date | Sep 29 2025 | Sep 5 2025 |
| Model type | — | — |
| Open source | No | Yes |
| Specifications | ||
Parameters | — | 1T |
Context window | — | 256k |
| Benchmarks | ||
Coding SWE-Bench VerifiedReal coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better. | 77.2%Best | 65.8% |
Science GPQA DiamondGraduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 83.4%Best | 75.1% |
Multimodal MMMUTests the AI on understanding images and text together across many college subjects. Higher is better. | 68% | — |
| Timeline | ||
| Release gap | Kimi K2 (0905) shipped 24 days before Claude Sonnet 4.5 | |
Claude Sonnet 4.5 leads Kimi K2 (0905) on 2 of the 2 benchmarks they both report (SWE-Bench Verified, GPQA Diamond). Kimi K2 (0905) shipped 24 days before Claude Sonnet 4.5, so benchmark comparisons should account for the intervening progress.
Kimi K2 (0905) is an open-source / open-weight model; Claude Sonnet 4.5 is proprietary.
On SWE-Bench Verified, Claude Sonnet 4.5 leads at 77.2% vs Kimi K2 (0905) at 65.8%. On GPQA Diamond, Claude Sonnet 4.5 leads at 83.4% vs Kimi K2 (0905) at 75.1%.