Command A vs Nova Premier vs Pixtral Large (25.02)
Nova Premier comes out ahead, 41 to 33 and 24 on our weighted score, though Pixtral Large (25.02) is 40% cheaper per token.
CohereCommand A
24/100- ECI—
- Price$2.50 / $10.00
- Context256K
- Our pick
AmazonNova Premier
41/100- ECI—
- Price$2.50 / $12.50
- Context1M
Mistral AIPixtral Large (25.02)
33/100- ECI—
- Price$2.00 / $6.00
- Context128K
Nova Premier is our pick
Nova Premier is the better all-round choice, scoring 41/100 against Pixtral Large (25.02) (33) and Command A (24). It leads on inputs & features and context window. Pixtral Large (25.02) wins on price. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
- CapabilityNot enough dataNo independent benchmark covers every model here yet
- Lowest pricePixtral Large (25.02)Pixtral Large (25.02) $3.00 · Command A $4.38 · Nova Premier $5.00 per 1M tokens (3:1 blend)
- Longest contextNova PremierNova Premier 1,000,000 · Command A 256,000 · Pixtral Large (25.02) 128,000 tokens
- Widest inputsNova PremierCommand A: Text · Nova Premier: Text, Images, PDFs, Video · Pixtral Large (25.02): Text, Images
- Self-hostingCommand APublishes downloadable weights
| Measure | Weight | Command A | Nova Premier | Pixtral Large (25.02) |
|---|---|---|---|---|
| Price | 50% | 19 | 17 | 27 |
| Inputs & features | 30% | 25 | 70 | 50 |
| Context window | 20% | 36 | 60 | 24 |
| Overall | 100% | 24/100 | 41/100 | 33/100 |
Left out because at least one model lacks the data: capability. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | Command ACohere | Nova PremierAmazon | Pixtral Large (25.02)Mistral AI |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $2.50 | $2.50 | $2.00 (best) |
| Output | $10.00 | $12.50 | $6.00 (best) |
| Cached input | — | $0.625 | — |
| Blended (3:1) | $4.38 | $5.00 | $3.00 (best) |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Official Cohere API | Official Amazon Bedrock API | Median of 3 providers |
| Limits | |||
| Context window | 256,000 tokens | 1,000,000 tokens (best) | 128,000 tokens |
| Max output | 8,000 tokens | 10,000 tokens (best) | 8,192 tokens |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | Yes | Yes |
| PDFs | No | Yes | No |
| Audio | No | No | No |
| Video | No | Yes | No |
| Reasoning | No | No | No |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | No |
| Availability | |||
| Weights | Open | Proprietary | Proprietary |
| API model ID | command-a-03-2025 | us.amazon.nova-premier-v1:0 | — |
| API providers | 3 (best) | 1 | 3 (best) |
| Released | Mar 13, 2025 | Apr 30, 2025 | Apr 8, 2025 |
| Knowledge cutoff | Jun 1, 2024 | Oct 2024 | — |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Command A$45.00
- Nova Premier$50.00
- Pixtral Large (25.02)$32.00
Which should you choose?
Which is better: Command A, Nova Premier or Pixtral Large (25.02)?
Nova Premier is the better all-round choice, scoring 41/100 against Pixtral Large (25.02) (33) and Command A (24). It leads on inputs & features and context window. Pixtral Large (25.02) wins on price. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, Command A, Nova Premier or Pixtral Large (25.02)?
Pixtral Large (25.02) is cheaper at $2.00 input / $6.00 output per million tokens (median across 3 API providers). Command A costs $2.50 input / $10.00 output per million tokens (official Cohere API price); Nova Premier costs $2.50 input / $12.50 output per million tokens (official Amazon Bedrock API price). At a typical mix of three input tokens to one output token, that is $3.00 per million tokens for Pixtral Large (25.02) versus $4.38 for Command A (1.5× as much) and $5.00 for Nova Premier (1.7× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Command A has not been scored yet, Nova Premier has not been scored yet and Pixtral Large (25.02) has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Command A, Nova Premier and Pixtral Large (25.02) yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. All three support tool calling for agent workflows.
Which has the bigger context window?
Nova Premier has the largest context window at 1,000,000 tokens, against 256,000 for Command A and 128,000 for Pixtral Large (25.02). Maximum output per response: Command A up to 8,000, Nova Premier up to 10,000, Pixtral Large (25.02) up to 8,192 tokens.
Which can read images, PDFs, audio or video?
Command A accepts text; Nova Premier accepts text, images, PDFs and video; Pixtral Large (25.02) accepts text and images. Nova Premier handles the widest range of inputs.
Are any of these open source?
Command A publishes its weights and can be self-hosted; Nova Premier and Pixtral Large (25.02) is proprietary.
Which is newer?
Nova Premier is the newest, released Apr 30, 2025. Pixtral Large (25.02) came out Apr 8, 2025; Command A came out Mar 13, 2025. Knowledge cutoff: Command A Jun 1, 2024, Nova Premier Oct 2024.
How do you decide the winner?
Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Oct 4, 2026.
