Cohere · language model
Command A: pricing, context and benchmarks
Command A is Cohere's model released on March 13, 2025. It costs $2.5 per million input tokens and $10 per million output tokens, with a 256K-token context window. No independent score has been published for it so far.
What you need to know about Command A
Cohere, based in Canada, put Command A online on March 13, 2025. It accepts text as input and produces text. Its context window reaches 256,000 tokens, larger than 41 % of the models in the registry, and it can write up to 8,192 tokens in one go.
Released more than 1 year ago, it belongs to a generation most vendors have since superseded.
Command A pricing
On OpenRouter, Command A costs $2.5 per million input tokens and $10 per million output tokens. Counting three tokens read for every token written, its average price is $4.375 per million, more expensive than 86 % of paid models (5.7 times the median of $0.76).
A concrete example: a task that sends 2,000 tokens (about 1,500 words) and receives a 500-token answer costs $0.01. At 1,000 tasks a day, budget $300 a month. The price calculator redoes this with your own volumes.
Performance and benchmarks
Epoch AI does not yet publish a capabilities index for Command A. We therefore show no made-up performance rating: only price, context and features count towards its score.
Features, privacy and local use
- It does not support tool calling: it is not suited to driving an AI agent that has to act.
- It can answer in an enforced JSON format (structured outputs).
- Its knowledge stops at August 2024.
Its weights are open and downloadable from Hugging Face: you can run it on your own machines without sending your data to a third party.
Other Cohere models
| Model | Released | Avg. price | ECI |
|---|---|---|---|
| Command A+ | Sep 22, 2026 | $0.6 | — |
| North Mini Code | Jun 17, 2026 | Free | — |
| Command R7B (12-2024) | Dec 14, 2024 | $0.07 | — |
| Command R (08-2024) | Aug 30, 2024 | $0.26 | — |
| Command R+ (08-2024) | Aug 30, 2024 | $4.38 | — |
Verdict
Computed score: 28/100, from public data (performance, price, context, features).
Strengths
- Open weights: self-hostable, your data stays with you.
Limitations
- Expensive: costlier than 86 % of paid models.
- No tool calling: unusable as an agent's brain.
- No independent score published: performance cannot be verified.
- Older generation, often replaced by a newer model from the same vendor.
- Text only: reads neither images nor files.
Avoid if…
- you process large volumes on a tight budget;
- you are building an agent that must call tools.
Alternatives
- o3$3.5
- Sonar Pro Search$6
- Sonar Pro$6
- Nano Banana Pro (Gemini 3 Pro Image)$4.5
- Aion-3.0$3.75
- GPT Audio$4.38
Comparisons and tools
Learn more
Frequently asked questions
How much does Command A cost?
$2.5 per million input tokens and $10 per million output tokens on OpenRouter (checked October 25, 2026). A task with 2,000 tokens read and 500 written costs $0.01.
How big is Command A's context window?
256,000 tokens, roughly 192,000 words, with a maximum of 8,192 output tokens.
Can Command A run locally?
Yes, its weights are published.
Is Command A suitable for an AI agent?
No: without tool calling it cannot act, only answer.
What are the alternatives to Command A?
The closest models on price and performance are o3, Sonar Pro Search, and Sonar Pro.
Related articles
- Mistral AI: models, pricing and agentsOctober 25, 2026
- LLM benchmarks: how to read model rankingsOctober 24, 2026
- Running a local LLM: hardware, software, and limitationsOctober 24, 2026
- Open-source or closed models for your AI agents?October 5, 2026
Neutrality. No affiliate links. No vendor pays to be reviewed, rated or ranked. Reviews are directed by François Chollet, neutral and factual, and every data point is dated and sourced. How we score