CohereCommand A 03
Best configuration ranks #436 of 610 on the BenchLeader Index at 43.5 ±6.8. Last measured 2 Sept 2026. Released 13 Mar 2025.
- Blended price
- $4.38/M
- $2.50 in · $10.00 out
- Output speed
- –
- First answer
- –
- Context
- 256k
- Overall index44
- Reasoning53
- Coding31
- Maths35
- Knowledge38
- Human preference53
Benchmark results
One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | default | Source |
|---|---|---|
| LMArena Hard Prompts | 1368#176 | LMArena |
| GPQA Diamond (Vals)not in index | 48.5%#118 | Vals AI |
Coding
| Benchmark | default | Source |
|---|---|---|
| LMArena Coding | 1389#182 | LMArena |
| LiveCodeBench | 35.1%#130 | Vals AI |
| SWE-bench (Vals)not in index | 7.8%#84 | Vals AI |
Maths
| Benchmark | default | Source |
|---|---|---|
| AIME (Vals) | 13.3%#84 | Vals AI |
| MGSM | 85.7%#64 | Vals AI |
Knowledge
Human preference
| Benchmark | default | Source |
|---|---|---|
| LMArena Text | 1354#175 | LMArena |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0040 | – | – |
| Summarise a 30-page report | 12,000 / 600 | $0.036 | – | – |
| Code edit | 6,000 / 1,500 | $0.030 | – | – |
| Agentic coding session | 60,000 / 4,000 | $0.190 | – | – |
| Structured extraction | 2,000 / 200 | $0.0070 | – | – |
See also
Data as of 9 Sept 2026. Compare with another model.