Command A: pricing, context window and benchmarks
Command A is a multimodal model from Cohere. It offers a 200,000-token context window, costs $13.30 per million input tokens and $67.41 per million output tokens, and averages 79.6 across the index's benchmark suite.
Pricing
Command A costs $13.30 per million input tokens and $67.41 per million output tokens. A blended 3:1 input-to-output mix works out to about $26.83 per million tokens. Cached input is listed at $1.13 per million tokens.
| Item | Price | Notes |
|---|---|---|
| Input | $13.30 | per 1M tokens |
| Output | $67.41 | per 1M tokens |
| Cached input | $1.13 | per 1M tokens |
| Blended 3:1 | $26.83 | estimate per 1M tokens |
| 1M in + 1M out | $80.71 | estimate |
Benchmarks
Scores below are modelled estimates on a 0–100 scale. Use them to compare relative position, not as official vendor results.
| Benchmark | Score | Rank | Percentile | Relative |
|---|---|---|---|---|
| MMLU-Pro | 80.3 | 173 | 71th | |
| GPQA Diamond | 74.2 | 161 | 73th | |
| SWE-bench Verified | 70.6 | 99 | 84th | |
| HumanEval | 83.1 | 159 | 74th | |
| MATH-500 | 84.2 | 120 | 80th | |
| AIME 2025 | 77.9 | 116 | 81th | |
| LiveCodeBench | 71.8 | 150 | 75th | |
| IFEval | 87.4 | 120 | 80th | |
| TruthfulQA | 81.2 | 147 | 76th | |
| HellaSwag | 88.3 | 130 | 78th | |
| MMMU | 76.4 | 127 | 79th | |
| Arena Elo | 1419.7 | 152 | 75th |
What Command A is good for
Command A is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $13.30 per million input tokens, it sits in the mid-range of the index, and the 200,000-token context window is enough for large files and multi-file diffs.
Strengths
- Large 200,000-token context window for long documents and agent traces.
- Accepts more than text, including image input where the provider supports it.
- Coding benchmarks are a relative strength within its tier.
Trade-offs
- Proprietary licence: you cannot download the weights or self-host the model.
- Premium input pricing at $13.30 per million tokens.
How it compares
| Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|
| Doubao 1.5 Pro | ByteDance | 200000 | $13.65 | 79.7 |
| Llama 4 Maverick Chat | Meta AI | 128000 | $5.09 | 79.7 |
| Sonar Pro Pro | Perplexity | 200000 | $13.59 | 79.7 |
| Command A | Cohere | 200000 | $13.30 | 79.6 |
| Nova Pro Pro | Amazon | 200000 | $13.03 | 79.5 |
| Qwen3-Coder | Alibaba Cloud | 200000 | $4.94 | 79.5 |
FAQ
How much does Command A cost?
Command A is listed at $13.30 per million input tokens and $67.41 per million output tokens. Cached input is $1.13 per million tokens. A blended 3:1 input-to-output mix is about $26.83 per million tokens.
What is the context window of Command A?
Command A has a context window of 200,000 tokens and can emit up to 16,384 output tokens per request in this index.
Is Command A open weights?
No. Command A is a proprietary model with a proprietary licence, so you cannot download the weights.
How does Command A score on benchmarks?
Command A averages 79.6 across the scored benchmarks in this index, ranking #132 of 600 models. Its strongest relative result is Arena Elo at 1419.7.
What is Command A best for?
Command A is tuned toward coding work. It is a reasonable candidate for code generation, review, debugging and repository-scale tasks, especially when paired with a retrieval layer. At $13.30 per million input tokens, it sits in the mid-range of the index, and the 200,000-token context window is enough for large files and multi-file diffs.