DeepSeek V4 Flash API

Use DeepSeek V4 Flash through the UncensoredAI OpenAI-compatible API with a single key and per-token billing. Point any OpenAI SDK at the endpoint, set the model to deepseek-v4-flash-0731, and stream or batch as you like.

Specifications

Model ID: deepseek-v4-flash-0731
Provider: DeepSeek
Context window: 1M
Max output: 66K
Input / 1M: $0.06
Cached / 1M: $0.0053
Output / 1M: $0.06

Example request

curl https://uncensoredai.dev/v1/chat/completions \
  -H "Authorization: Bearer sk-***" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4-flash-0731",
    "messages": [{ "role": "user", "content": "Hello" }],
    "stream": true
  }'

Get an API key Full pricing Try it in the playground