Command A
Cohere's enterprise-ready flagship, built for agents, RAG, tool use, multilingual work, and structured outputs.
Pricing
Input$2.5 / 1M checked Jul 19, 2026
Cached inputNot surfaced in model page
Output$10 / 1M checked Jul 19, 2026
Priced as a premium enterprise model.
Cost bandPremium
Free tierProduction is paid.
Limits
Context window256k checked Jul 19, 2026
Max output8k
Knowledge cutoffJune 1, 2024
DeploymentCohere API
OpenAI-compatible APINo
ModalitiesText, Image input
Where it fits
Best forEnterprise agents, multilingual workflows, and retrieval-heavy product experiences.
Business fitWorth considering when enterprise packaging and multilingual support matter more than benchmark hype.
Capabilities
ReasoningTool useCitationsStructured outputsMultilingualSafety modes
Benchmark scores
Composite77 / 100
Coding72 / 100 (HumanEval)
Reasoning76 / 100 (MMLU-Pro)
Long context81 / 100 (RULER 256K)
Vision65 / 100
Instruction following80 / 100 (IFEval)
Editorial aggregate of published benchmark results as of Mar 1, 2025. These are not independent measurements run by this site.
Change history
- Mar 13, 2025New modelCohere Command A released Enterprise flagship: 256k context, multilingual RAG, citations, tool use at $2.5/$10 per 1M.
Data
Last checkedJul 19, 2026
Cross-checked againstopenrouter.ai/cohere/command-a
Verify with provider documentation before committing spend. Report an error and it gets fixed with a logged correction.