DeepSeek V3.1 is 10.4× cheaper per million tokens (blended). Command A has the larger context window (256,000 vs 163,840 tokens, 1.6×).

Command A vs DeepSeek V3.1

Live catalog fields. Best-in-row is highlighted.

Command Aopen

Cohere · cohere__command-a

DeepSeek V3.1open

DeepSeek · deepseek__deepseek-chat-v3.1

Identity

Fieldcohere__command-adeepseek__deepseek-chat-v3.1
ProviderCohereDeepSeek
Slugcohere__command-adeepseek__deepseek-chat-v3.1
Statuslivelive
Open weightsYesYes
License
HuggingFaceCohereForAI/c4ai-command-a-03-2025deepseek-ai/DeepSeek-V3.1

Quality

Fieldcohere__command-adeepseek__deepseek-chat-v3.1
Quality index (0–100)

Cost

Fieldcohere__command-adeepseek__deepseek-chat-v3.1
Input $ / 1M tokens$2.50$0.25
Cached input $ / 1M$0.13
Output $ / 1M tokens$10.00$0.95

Context

Fieldcohere__command-adeepseek__deepseek-chat-v3.1
Max context tokens256,000163,840
Max output tokens8,19232,768

Modalities

Fieldcohere__command-adeepseek__deepseek-chat-v3.1
Inputtexttext
Outputtexttext