Cohere
Command A is a Cohere model profile, generally aimed at enterprise language, retrieval, multilingual, and tool-using workflows. The benchmark data on this page helps compare the model's practical quality, runtime, and price against both frontier chat models and other enterprise-oriented alternatives.
Cohere model releasesQueryable facts extracted from the upstream model payload.
Rank #28 across 293 models.
Rank #303 across 323 models.
Rank #277 across 325 models.
Rank #227 across 269 models.
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Higher bars mean stronger relative placement.
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 13.5 | #352 |
| Artificial Analysis Coding Index | coding | 9.9 | #318 |
| Artificial Analysis Math Index | math | 13.0 | #227 |
| MMLU-Pro | reasoning | 71.2% | #213 |
| reasoning |
| 52.7% |
| #322 |
| Humanity's Last Exam | reasoning | 4.6% | #336 |
| LiveCodeBench | coding | 28.7% | #234 |
| SciCode | coding, reasoning | 28.1% | #274 |
| MATH-500 | math | 81.9% | #104 |
| AIME | math | 9.7% | #134 |
| Output Speed | speed | 50.7 tok/s | #223 |
| Time to First Token | speed | 0.44s | #28 |
| Blended Price | cost | $4.38/M | #271 |
| Input Price | cost | $2.50/M | #277 |
| Output Price | cost | $10.00/M | #257 |
| Value Index | cost, overall | 3.1 | #303 |