2 vs 0 benchmarks won
Google Gemini 3.1 Pro | OpenAI GPT-5.4 | |
|---|---|---|
| Overview | ||
| Company | OpenAI | |
| Release date | Feb 19 2026 | Mar 5 2026 |
| Model type | — | — |
| Open source | No | No |
| Specifications | ||
Parameters | — | — |
Context window | — | — |
| Benchmarks | ||
Science reasoning GPQA Diamond | 94.3% | 92.8% |
Software engineering SWE-Bench Verified | 80.6% | — |
Multimodal understanding MMMU | — | — |
| Timeline | ||
| Release gap | Gemini 3.1 Pro shipped 14 days before GPT-5.4 | |
Gemini 3.1 Pro leads GPT-5.4 on 2 of the tracked benchmarks (GPQA Diamond, SWE-Bench Verified, MMMU). Gemini 3.1 Pro shipped 14 days before GPT-5.4, 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, Gemini 3.1 Pro scores 94.3%, 1.5 points above GPT-5.4 at 92.8%.