Back to models
DeepSeekactive

DeepSeek v4

DeepSeek-V4 Preview (Flash/Pro variants) is a Mixture-of-Experts model series with 1M token context, top-tier reasoning, and strong agent capabilities. Fully OpenAI/Anthropic compatible.

Input / 1M tokens

$0.140

Output / 1M tokens

$0.280

Context window

1.0M

Capabilities

  • Streaming
  • Function Calling