0 vs 1 benchmarks won
DeepSeek DeepSeek-V4-Pro | OpenAI GPT-5.2 | |
|---|---|---|
| Overview | ||
| Company | DeepSeek | OpenAI |
| Release date | Apr 24 2026 | Dec 11 2025 |
| Model type | — | — |
| Open source | Yes | No |
| Specifications | ||
Parameters | — | — |
Context window | — | — |
| 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. | — | 80% |
Science GPQA DiamondGraduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better. | 90.1% | 92.4%Best |
| Timeline | ||
| Release gap | GPT-5.2 shipped 134 days before DeepSeek-V4-Pro | |
GPT-5.2 leads DeepSeek-V4-Pro on 1 of the 1 benchmark they both report (GPQA Diamond). GPT-5.2 shipped 134 days before DeepSeek-V4-Pro, so benchmark comparisons should account for the intervening progress.
DeepSeek-V4-Pro is an open-source / open-weight model; GPT-5.2 is proprietary.
On GPQA Diamond, GPT-5.2 leads at 92.4% vs DeepSeek-V4-Pro at 90.1%.