n8n-nodes-nvidia-nim-chat

Use NVIDIA NIM inference microservices as the chat language model in n8n AI Agent and AI Chain workflows. The node speaks the OpenAI-compatible API surface that NIM exposes, so you can point it at NVIDIA-hosted endpoints or your own NIM deployment.

n8n is a fair-code licensed workflow automation platform.

Contents

• Installation
• Quick start
• What this node does
• Credentials
• Compatibility
• Developing this package
• Resources
• Version history

Installation

1. In n8n, open Settings → Community nodes (or your admin's equivalent).
2. Install the package n8n-nodes-nvidia-nim-chat.

For detailed steps, security notes, and Docker or CLI options, see the official guide: Installing community nodes.

Quick start

1. Create credentials
   In n8n, add NVIDIA NIM Chat API credentials (see Credentials).

2. Add the node
   Add NVIDIA NIM Chat Model to your workflow.

3. Connect it
   Wire the node's Model output to the Chat Model input of an AI Agent or AI Chain node (this node has no regular inputs; it only supplies the model).

4. Pick a model
   The Model dropdown is loaded from your GET /models endpoint. Choose the model your NIM deployment exposes.

5. Optional tuning
   Open Options for temperature, top-p, max tokens, retries, timeouts, and response format (text or JSON).

If you are new to n8n, the Try it out docs help you run your first workflows.

What this node does

| Aspect | Details | |--------|---------| | Role | Supplies a chat LLM to AI Agent / AI Chain (output type: AI language model). | | API | OpenAI-compatible chat completions against your configured Base URL. | | Models | List is fetched dynamically from the API so it matches your account or self-hosted NIM. |

JSON responses: If you set Response Format to JSON, include the word json in your prompt (as noted in the node UI) so the model returns valid JSON.

Credentials

Create NVIDIA NIM Chat API credentials in n8n:

| Field | Description | |--------|-------------| | API Key | Your NVIDIA NGC API key. Create or manage keys at NVIDIA NGC. Stored and sent as a Bearer token in the Authorization header. | | Base URL | Root URL for the OpenAI-compatible API. Default: https://integrate.api.nvidia.com/v1. Use your own base URL if you run NIM on-premises or in your cloud. |

Use Test on the credential to verify access; it calls GET /models on your base URL.

Compatibility

• n8n: Use a version that supports community nodes and built-in AI Agent / AI Chain (and LangChain-style model inputs). This package targets n8n node API version 1.
• A specific minimum n8n version is not pinned in this repo; if something fails to appear in the editor, upgrade n8n first.

Developing this package

Clone the repository, install dependencies, then use the n8n node CLI:

npm install
npm run build    # production build → dist/
npm run dev      # local n8n with this package (see n8n-node dev docs)
npm run lint     # eslint via n8n-node

Release/version notes live in CHANGELOG.md.

Resources

• NVIDIA NIM documentation
• n8n community nodes
• Creating nodes overview
• Repository

Version history

See CHANGELOG.md for release history.