DeepSeek/deepseek-v4-pro/Released Apr 2026

DeepSeek V4 Pro

Large MoE for reasoning and coding

Text

About DeepSeek V4 Pro

DeepSeek V4 Pro is the top-tier model of the V4 family, targeting advanced reasoning, coding, and complex multi-step tasks. The Mixture-of-Experts architecture activates 49B parameters per token out of a 1.6T total, and the model supports structured outputs and tool use natively.

MetricValue
Parameter Count1.6 trillion (49 billion active)
Mixture of ExpertsYes
Active Parameter Count49 billion
Context Length1,048,576 tokens
MultilingualYes
Tool UseYes
Structured OutputsYes

Ready to build with DeepSeek V4 Pro?

Try DeepSeek V4 Pro in the Workbench to prompt it, compare outputs, and iterate on prompts without writing any code. When you're ready to ship, call the same model from our API and build your own apps on top of it.

curl -sSf -X POST https://hub.oxen.ai/api/ai/chat/completions \
    -H "Content-Type: application/json" \
    -H "Authorization: Bearer $OXEN_API_KEY" \
    -d '{
  "model": "deepseek-v4-pro",
  "messages": [
    {
      "role": "user",
      "content": "Try sending a message."
    }
  ]
}'

API endpoint

See the API reference for request and response formats.

POSThttps://hub.oxen.ai/api/ai/chat/completions