---
id: "cohere/command-a"
also: ["cohere/command-a-03-2025", "CohereForAI/c4ai-command-a-03-2025", "vercel/cohere/command-a"]
card_usd_per_m: "2.5 in, 10.0 out, at cohere"
cheapest_list_usd_per_m: "2.5 in, 10.0 out, at cohere on openrouter"
cheapest_effective_usd_per_m: "2.4997 in, 10.0 out, at cohere on openrouter"
context: 256000
max_output: 8192
input: ["text"]
output: ["text"]
released: "2025-03-13"
api: ["/api/v1/chat/completions", "/api/v1/completions", "/api/v1/messages", "/api/v1/responses"]
params: ["frequency_penalty", "max_tokens", "presence_penalty", "response_format", "seed", "stop", "structured_outputs", "temperature", "tool_choice", "tools", "top_k", "top_p"]
endpoints: 2
gateways: "openrouter (1), vercel (1)"
no_training: 1
trains: 0
retains_prompts: 1
---
# Cohere: Command A

Command A is Cohere's most performant model to date, excelling at tool use, agents, retrieval augmented generation (RAG), and multilingual use cases. Command A has a context length of 256K, only requires two GPUs to run, and has 150% higher throughput compared to Command R+ 08-2024.