Command A Reasoning
Cohere's reasoning variant for complex agentic tasks, free until rate limits are reached.
Pricing
InputFree until rate limits
Cached inputNot surfaced
OutputFree until rate limits
Model page explicitly states free use until rate limits are reached.
Cost bandFree-to-eval
Free tierFree until rate limits are reached; production usage routes through stricter access paths.
Limits
Context window256k
Max output32k
Knowledge cutoffJune 1, 2024
DeploymentCohere API
OpenAI-compatible APINo
ModalitiesText, Image input
Where it fits
Best forEvaluation, research, and agentic tasks where you want Cohere's stack and free experimentation first.
Business fitInteresting for evaluation and prototyping, but production access is more constrained.
Capabilities
ReasoningTool useCitationsStructured outputsMultilingualSafety modes
Benchmark scores
Composite81 / 100
Coding78 / 100 (HumanEval (vendor))
Reasoning91 / 100 (GPQA Diamond (vendor))
Long context78 / 100 (RULER (vendor))
Vision62 / 100
Instruction following82 / 100 (IFEval (vendor))
Editorial aggregate of published benchmark results as of Sep 1, 2025. These are not independent measurements run by this site.
Change history
No recorded changes for this model yet. Pricing and context are checked daily; changes appear here with dates and sources.
Data
Last checkedMar 31, 2026
Verify with provider documentation before committing spend. Report an error and it gets fixed with a logged correction.