Command A Max vs Command A Pro: which model should you choose?
We compared Command A Max and Command A Pro on price, context window and 12 benchmarks. Command A Pro has the stronger average benchmark score; Command A Pro is cheaper on a blended 3:1 basis.
Command A Max
Command A Max is a multimodal model from Cohere. It offers a 400,000-token context window, costs $14.28 per million input tokens and $70.22 per million output tokens, and averages 79.9 across the index's benchmark suite.
- Input
- $14.28 / 1M
- Output
- $70.22 / 1M
- Context
- 400,000
- Avg score
- 79.9
- Open weights
- No
Command A Pro
Command A Pro is a multimodal model from Cohere. It offers a 200,000-token context window, costs $14.28 per million input tokens and $68.05 per million output tokens, and averages 81.3 across the index's benchmark suite.
- Input
- $14.28 / 1M
- Output
- $68.05 / 1M
- Context
- 200,000
- Avg score
- 81.3
- Open weights
- No
Specification comparison
| Spec | Command A Max | Command A Pro | Edge |
|---|---|---|---|
| Input price / 1M | $14.28 | $14.28 | Tie |
| Output price / 1M | $70.22 | $68.05 | B |
| Context window | 400,000 | 200,000 | A |
| Max output | 8,192 | 16,384 | B |
| Average score | 79.9 | 81.3 | B |
| Arena Elo | 1443.7 | 1434.6 | A |
| Open weights | No | No | Tie |
Benchmark comparison
Command A Max wins 3, Command A Pro wins 9, and 0 benchmarks are tied. The table shows the modelled score for each model and the gap.
| Benchmark | Command A Max | Command A Pro | Gap | Winner |
|---|---|---|---|---|
| MMLU-Pro | 84.3 | 81.3 | 3.0 | Command A Max |
| GPQA Diamond | 74.5 | 79.1 | 4.6 | Command A Pro |
| SWE-bench Verified | 69.7 | 71.2 | 1.5 | Command A Pro |
| HumanEval | 83.4 | 87.5 | 4.1 | Command A Pro |
| MATH-500 | 85.7 | 83.1 | 2.6 | Command A Max |
| AIME 2025 | 77.2 | 78.2 | 1.0 | Command A Pro |
| LiveCodeBench | 73.1 | 74.0 | 0.9 | Command A Pro |
| IFEval | 86.7 | 89.0 | 2.3 | Command A Pro |
| TruthfulQA | 82.2 | 83.3 | 1.1 | Command A Pro |
| HellaSwag | 87.1 | 90.1 | 3.0 | Command A Pro |
| MMMU | 74.6 | 77.2 | 2.6 | Command A Pro |
| Arena Elo | 1443.7 | 1434.6 | 9.1 | Command A Max |
Price analysis
On a blended 3:1 input-to-output basis, Command A Max costs about $28.27 per million tokens and Command A Pro costs about $27.72. Command A Pro is roughly 2% cheaper than Command A Max. At high volume, that gap compounds quickly, so price should carry real weight unless the quality difference is decisive.
Command A Max has the larger context window at 400,000 tokens, compared with 200,000 for Command A Pro. That matters for long documents, large codebases and multi-step agent traces.
Verdict
Across 12 shared benchmarks, Command A Max wins 3, Command A Pro wins 9, and 0 are tied. On the index average, Command A Pro leads Command A Max by 1.4 points (81.3 versus 79.9).
The price picture favours Command A Pro. The context picture favours Command A Max. Those two dimensions often matter more than a few benchmark points once a model is good enough for the task.
If you need a single default, start with Command A Pro for quality-sensitive work, and check whether its price holds at your expected volume before committing.
Choose Command A Max if
- You need the larger 400,000-token context window.
Choose Command A Pro if
- You want the higher average benchmark score in this index.
- Unit cost matters and you want the cheaper blended price.
- You need the larger 16,384-token output limit.
FAQ
Is Command A Max better than Command A Pro?
On the index average, Command A Pro leads Command A Max by 1.4 points (81.3 versus 79.9). The better choice depends on price, context and your task. Command A Max wins 3 benchmarks, Command A Pro wins 9.
Which is cheaper, Command A Max or Command A Pro?
Command A Pro is cheaper on a blended 3:1 basis at about $27.72 per million tokens, versus $28.27 for Command A Max.
Which has the larger context window?
Command A Max has the larger context window at 400,000 tokens. Command A Pro has 200,000 tokens.
Are these models open weights?
Command A Max is proprietary; Command A Pro is proprietary.
Which should I choose for production?
If you need a single default, start with Command A Pro for quality-sensitive work, and check whether its price holds at your expected volume before committing.