1 vs 1 benchmarks won
Moonshot AI Kimi K2.5 | OpenAI GPT-5.4 | |
|---|---|---|
| Overview | ||
| Company | Moonshot AI | OpenAI |
| Release date | Jan 27 2026 | Mar 5 2026 |
| Model type | — | — |
| Open source | Yes | No |
| Specifications | ||
Parameters | 1T | — |
Context window | 256k | — |
| Benchmarks | ||
Science reasoning GPQA Diamond | 87.6% | 92.8% |
Software engineering SWE-Bench Verified | 76.8% | — |
Multimodal understanding MMMU | — | — |
| Timeline | ||
| Release gap | Kimi K2.5 shipped 37 days before GPT-5.4 | |
Kimi K2.5 and GPT-5.4 are evenly matched across the benchmarks they both publish. Kimi K2.5 shipped 37 days before GPT-5.4, so benchmark comparisons should account for the intervening progress.
Kimi K2.5 is an open-source / open-weight model; GPT-5.4 is proprietary.
On GPQA Diamond, GPT-5.4 scores 92.8%, 5.2 points above Kimi K2.5 at 87.6%.