Discover And Chat
A 3-step recipe with NVIDIA NIM. We can help you get access.
Recipe · 3 steps
Discover And Chat
with NVIDIA NIM
Discover a served model and run a chat completion against it.
- List the models exposed by this NIM endpoint to discover the exact model strings that can be passed to chat completions.
- Retrieve metadata for the requested model id to confirm the endpoint actually serves it before attempting inference.
- Send a chat completion request using the confirmed model id and the supplied user prompt.
What you'll need to have ready
- apiKey: NVIDIA developer API key (nvapi-...) sent as a Bearer token.
- modelId: The model id to confirm and use (e.g. meta/llama-3.3-70b-instruct).
- prompt: The user prompt to send to the chat model.
- maxTokens: Maximum number of tokens to generate.
For your developer
Arazzo workflow nvidia-nim:discover-and-chat, 3 operations: listModels → getModel → createChatCompletion. Specs on APIs.io.