Command A
Cohere · released Mar 13, 2025
Command A is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and coding use cases. Compared to other leading proprietary...
Specification
- Context window
- 256K
- Max output
- 8K
- Knowledge cutoff
- Aug 31, 2024
- Parameters
- 111.1B
- Licence
- cc-by-nc-4.0
- Serving providers
- 1
- Moderated
- Yes
Intelligence
7.5
25th percentile
Coding
—
Coding Index
Agentic
9.2
Agentic Index
Output speed
32 t/s
Median across providers
Latency
332ms
Time to first token
Cost per task
$0.38
Estimated
Benchmarks
Where the score comes from
The Intelligence Index is a composite. These are the underlying evaluations this model was actually measured on.
Evaluation scores
Percentage correct · higher is better
- MMLU-Pro71.2%
- GPQA Diamond52.7%
- IFBench36.5%
- LiveCodeBench28.7%
- SciCode28.1%
- AA-LCR (long context)20.0%
- τ²-bench (Telecom)15.2%
- AIME 202513.0%
- Humanity's Last Exam4.0%
- Terminal-Bench Hard0.8%
An evaluation missing from this list was not run for this model — it is not a zero.
View as table
| Evaluation | Score |
|---|---|
| MMLU-Pro | 71.2% |
| GPQA Diamond | 52.7% |
| IFBench | 36.5% |
| LiveCodeBench | 28.7% |
| SciCode | 28.1% |
| AA-LCR (long context) | 20.0% |
| τ²-bench (Telecom) | 15.2% |
| AIME 2025 | 13.0% |
| Humanity's Last Exam | 4.0% |
| Terminal-Bench Hard | 0.8% |
Against its peers
Intelligence Index · this model highlighted, nearest peers in grey
- 8.2
- GPT-4 TurboOpenAI7.7
- Command ACohere7.5
- Nemotron 3 Nano 30B A3BNVIDIA7.2
- Mistral Large 2407Mistral AI7.0
- 6.9
- GLM 4.5VZ.ai6.8
- GPT-4OpenAI6.8
Peers are the models sitting closest on the Intelligence Index — the set you would realistically choose between.
View as table
| Model | Intelligence |
|---|---|
| Qwen3 VL 8B Instruct | 8.2 |
| GPT-4 Turbo | 7.7 |
| Command A | 7.5 |
| Nemotron 3 Nano 30B A3B | 7.2 |
| Mistral Large 2407 | 7.0 |
| Qwen2.5 Coder 32B Instruct | 6.9 |
| GLM 4.5V | 6.8 |
| GPT-4 | 6.8 |
Percentile among all indexed models
Pricing
What it costs to run
List prices per million tokens, plus what one representative task works out to.
List price
- Input / 1M tokens
- $2.5
- Output / 1M tokens
- $10
- Cached input / 1M
- Not offered
- Blended 3:1
- $4.38
One task, estimated
$0.38
- Input tokens
- 50,000
- Output tokens
- 25,000
- Profile
- Standard
Estimated from list pricing: 50K input tokens plus 80K output tokens for reasoning models (25K for non-reasoning).
Serving providers
Speed and latency figures are medians across these providers, so a widely-served model reports a blend rather than any single endpoint.