Cohere · AI model
Command A Reasoning
Cohere's first reasoning model (111B parameters, August 2025) for agentic and tool-use tasks in 23 languages; 256K context, 32K output; thinking is on by default and can be disabled or capped with a token budget.
Last verified Available 5 sources
Key specs
- Provider
- Cohere
- Family
- Command A
- API identifier
- command-a-reasoning-08-2025
- Availability
- Available
- Release date
- 2025-08-21
- License
- Open weights (CC-BY-NC 4.0 License with Acceptable Use Addendum)
- Context window
- 256,000 tokens
- Max output
- 32,000 tokens
- Knowledge cutoff
- June 2024
- API access
- Yes
- Consumer app
- No
Capabilities
Inputs
- Text input Yes
- Image input No Image Inputs marked as not supported in the capabilities panel on the model's docs page; docs model table lists modality as Text
- Audio input No Docs model table lists modality as Text only
- Video input No Docs model table lists modality as Text only
- File input Unknown
Outputs
- Text output Yes
- Image output No Docs model table lists modality as Text only
- Audio output No Docs model table lists modality as Text only
Features
- Reasoning Optional Thinking enabled by default; can be disabled or limited with token_budget via the thinking parameter
- Tool calling Yes
- Web search Unknown
"Unknown" means we have not verified this capability yet; it does not mean the model lacks it.
Pricing
No current prices listed.
Compare Command A Reasoning
Guides
About Command A Reasoning
Command A Reasoning is a 111B-parameter hybrid reasoning model released on August 21, 2025. Thinking is enabled by default and can be disabled or limited with a token budget through the Chat API's thinking parameter. It targets tool use, agentic workflows and multilingual reasoning in 23 languages, and weights are published under a non-commercial CC-BY-NC 4.0 licence. On the Cohere API it is free up to the rate limits, and production use runs through Model Vault.
Strengths and weaknesses
Strengths
- 256,000-token context window and up to 32,000 output tokens
- Reasoning can be switched off or capped with a thinking token budget
- Documented for tool use and agentic workflows
- Supports 23 languages
Weaknesses
- No per-token API price is published; production keys are limited like trial keys for this model, and production use requires Model Vault or contacting sales
- Text-only input
- Cohere lists 4x H100 GPUs for production self-hosting, versus two GPUs for Command A
Sources
Last verified Page updated 2026-10-13