Command A Max vs Kimi K1.5 Instruct: which model should you choose?
We compared Command A Max and Kimi K1.5 Instruct on price, context window and 12 benchmarks. Kimi K1.5 Instruct has the stronger average benchmark score; Kimi K1.5 Instruct 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
Kimi K1.5 Instruct
Kimi K1.5 Instruct is a multimodal model from Moonshot AI. It offers a 200,000-token context window, costs $5.15 per million input tokens and $28.20 per million output tokens, and averages 80.5 across the index's benchmark suite.
- Input
- $5.15 / 1M
- Output
- $28.20 / 1M
- Context
- 200,000
- Avg score
- 80.5
- Open weights
- Yes
Specification comparison
| Spec | Command A Max | Kimi K1.5 Instruct | Edge |
|---|---|---|---|
| Input price / 1M | $14.28 | $5.15 | B |
| Output price / 1M | $70.22 | $28.20 | B |
| Context window | 400,000 | 200,000 | A |
| Max output | 8,192 | 4,096 | A |
| Average score | 79.9 | 80.5 | B |
| Arena Elo | 1443.7 | 1431.8 | A |
| Open weights | No | Yes | B |
Benchmark comparison
Command A Max wins 6, Kimi K1.5 Instruct wins 6, and 0 benchmarks are tied. The table shows the modelled score for each model and the gap.
| Benchmark | Command A Max | Kimi K1.5 Instruct | Gap | Winner |
|---|---|---|---|---|
| MMLU-Pro | 84.3 | 83.0 | 1.3 | Command A Max |
| GPQA Diamond | 74.5 | 77.4 | 2.9 | Kimi K1.5 Instruct |
| SWE-bench Verified | 69.7 | 67.8 | 1.9 | Command A Max |
| HumanEval | 83.4 | 84.0 | 0.6 | Kimi K1.5 Instruct |
| MATH-500 | 85.7 | 82.6 | 3.1 | Command A Max |
| AIME 2025 | 77.2 | 80.4 | 3.2 | Kimi K1.5 Instruct |
| LiveCodeBench | 73.1 | 71.1 | 2.0 | Command A Max |
| IFEval | 86.7 | 86.2 | 0.5 | Command A Max |
| TruthfulQA | 82.2 | 84.3 | 2.1 | Kimi K1.5 Instruct |
| HellaSwag | 87.1 | 90.2 | 3.1 | Kimi K1.5 Instruct |
| MMMU | 74.6 | 78.6 | 4.0 | Kimi K1.5 Instruct |
| Arena Elo | 1443.7 | 1431.8 | 11.9 | 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 Kimi K1.5 Instruct costs about $10.91. Kimi K1.5 Instruct is roughly 61% 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 Kimi K1.5 Instruct. That matters for long documents, large codebases and multi-step agent traces.
Verdict
Across 12 shared benchmarks, Command A Max wins 6, Kimi K1.5 Instruct wins 6, and 0 are tied. On the index average, Kimi K1.5 Instruct leads Command A Max by 0.6 points (80.5 versus 79.9).
The price picture favours Kimi K1.5 Instruct. 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 Kimi K1.5 Instruct 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.
- You need the larger 8,192-token output limit.
Choose Kimi K1.5 Instruct if
- You want the higher average benchmark score in this index.
- Unit cost matters and you want the cheaper blended price.
- You need open weights for self-hosting or fine-tuning.
FAQ
Is Command A Max better than Kimi K1.5 Instruct?
On the index average, Kimi K1.5 Instruct leads Command A Max by 0.6 points (80.5 versus 79.9). The better choice depends on price, context and your task. Command A Max wins 6 benchmarks, Kimi K1.5 Instruct wins 6.
Which is cheaper, Command A Max or Kimi K1.5 Instruct?
Kimi K1.5 Instruct is cheaper on a blended 3:1 basis at about $10.91 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. Kimi K1.5 Instruct has 200,000 tokens.
Are these models open weights?
Command A Max is proprietary; Kimi K1.5 Instruct is open weights.
Which should I choose for production?
If you need a single default, start with Kimi K1.5 Instruct for quality-sensitive work, and check whether its price holds at your expected volume before committing.