DeepSeek
DeepSeek chat, Responses and text completions with streaming.
Use DeepSeek models through DeepintShield. The gateway sends your DeepSeek API key with bearer authentication to the default upstream.
What works
- List models
- Chat Completions, Responses and text completions, all with SSE streaming
- Function tools, tool choice, structured output and thinking controls
Limits to know: Other operations are rejected locally. The stored Responses lifecycle is not enabled. deepseek-v4-pro is not supported on Responses, so use Chat Completions for it.
How to connect
Add your DeepSeek API key in Model Hub → AI Providers, create a Virtual Key, then call deepseek/<model> through https://app.deepintshield.com/v1 with any OpenAI-compatible client.
Read the DeepSeek provider docs. Requests get the same Virtual Key budgets, guardrails and logs as every other provider on the AI gateway.