1 vs 1 benchmarks won
Anthropic Claude 3 Opus | DeepSeek DeepSeek-V4-Pro | |
|---|---|---|
| Overview | ||
| Company | Anthropic | DeepSeek |
| Release date | Mar 4 2024 | Apr 24 2026 |
| Model type | — | — |
| Open source | No | Yes |
| Specifications | ||
Parameters | — | — |
Context window | — | — |
| Benchmarks | ||
Science reasoning GPQA Diamond | 50.4% | 90.1% |
Software engineering SWE-Bench Verified | 33% | — |
Multimodal understanding MMMU | — | — |
| Timeline | ||
| Release gap | Claude 3 Opus shipped 781 days before DeepSeek-V4-Pro | |
Claude 3 Opus and DeepSeek-V4-Pro are evenly matched across the benchmarks they both publish. Claude 3 Opus shipped 781 days before DeepSeek-V4-Pro, so benchmark comparisons should account for the intervening progress.
DeepSeek-V4-Pro is an open-source / open-weight model; Claude 3 Opus is proprietary.
On GPQA Diamond, DeepSeek-V4-Pro scores 90.1%, 39.7 points above Claude 3 Opus at 50.4%.