DeepSeek

DeepSeek V4 Pro

Large-scale Mixture-of-Experts model with a 1M-token context.

ReasoningLong contextValue

$ / 1M input

$0.435

What you pay per million prompt tokens.

$ / 1M output

$0.870

What you pay per million completion tokens.

Blended (70 / 30)

$0.566

Typical chat workload mix.

Context window

1,048,576 tokens