0 vs 2 benchmarks won
DeepSeek DeepSeek V3 Base | Google Gemini 3.1 Pro | |
|---|---|---|
| Overview | ||
| Company | DeepSeek | |
| Release date | Dec 2024 | Feb 19 2026 |
| Model type | — | — |
| Open source | Yes | No |
| Specifications | ||
Parameters | — | — |
Context window | — | — |
| Benchmarks | ||
Science reasoning GPQA Diamond | — | 94.3% |
Software engineering SWE-Bench Verified | — | 80.6% |
Multimodal understanding MMMU | — | — |
| Timeline | ||
| Release gap | DeepSeek V3 Base shipped 445 days before Gemini 3.1 Pro | |
Gemini 3.1 Pro leads DeepSeek V3 Base on 2 of the tracked benchmarks (GPQA Diamond, SWE-Bench Verified, MMMU). DeepSeek V3 Base shipped 445 days before Gemini 3.1 Pro, so benchmark comparisons should account for the intervening progress.
DeepSeek V3 Base is an open-source / open-weight model; Gemini 3.1 Pro is proprietary.
Direct benchmark comparisons are unavailable — at least one of these models has not published scores on GPQA Diamond, SWE-Bench Verified, or MMMU.