DeepSeek
DeepSeek-V3 is the latest model from the DeepSeek team, building upon the instruction following and coding abilities of the previous versions. Pre-trained on nearly 15 trillion tokens, the reported evaluations...
DeepSeek release notesRank #295 across 566
Rank #156 across 222
Rank #114 across 371
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Higher bars mean stronger relative placement.
Streaming speed is not measured for this model yet.
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 14.2 | #295 |
| Artificial Analysis Coding Index | coding | 23.0 | #156 |
| Artificial Analysis Math Index | math | 26.0 | #195 |
| MMLU-Pro | reasoning | 75.2% | #180 |
| reasoning |
| 55.7% |
| #364 |
| Humanity's Last Exam | reasoning | 3.6% | #510 |
| LiveCodeBench | coding | 35.9% | #194 |
| SciCode | coding, reasoning | 35.4% | #237 |
| MATH-500 | math | 88.7% | #80 |
| AIME | math | 25.3% | #93 |
| Blended Price | cost | $0.493/M | #114 |
| Input Price | cost | $0.360/M | #147 |
| Output Price | cost | $0.890/M | #111 |
| Value Index | cost, overall | 28.8 | #121 |