Groq
Groq-hosted models through Chat Completions and Responses.
Use models hosted on Groq through DeepintShield, with your Groq key stored on the gateway.
What works
- Chat Completions, and the Responses API through chat
- List models
- Tool calling and
reasoning_effort
Limits to know: No embeddings, image generation, files or batch, and no image, audio or file inputs. Text completions are only available through the LiteLLM-compatibility fallback, which is off by default and not recommended.
How to connect
Add your Groq API key in Model Hub → AI Providers, create a Virtual Key, then call groq/<model> through https://app.deepintshield.com/v1 with any OpenAI-compatible client.
Read the Groq provider docs. Requests get the same Virtual Key budgets, guardrails and logs as every other provider on the AI gateway.