DeepSeek V3
Benchmark scores
| Benchmark | Category | Rank | Score | Captured |
|---|---|---|---|---|
| OpenRouter · Weekly Usage | usage | #4 | #10 | 2026-06-09 |
| Aider Polyglot | code | #5 | 74.2% | 2025-10-03 |
| Chatbot Arena · Open Weights | chat | #7 | 1424 | 2026-07-30 |
| SWE-bench Verified | agents | #14 | 70.0% | 2026-02-17 |
| AA Coding Index | code | #16 | 44.2% | 2026-07-31 |
| Terminal-Bench 2.0 | agents | #19 | 39.6% | 2026-02-08 |
| Chatbot Arena | chat | #24 | 1424 | 2026-07-30 |