n8n-nodes-nvidia-nim

This is an n8n community node that integrates NVIDIA NIM (NVIDIA Inference Microservices) as a first-class Chat Model (ai_languageModel connection type) in advanced AI workflows.

It is designed to connect directly into AI Agents or the Model Selector nodes in n8n, enabling both cloud-hosted and self-hosted model execution.

Features

• Dynamic Model Loading: Queries the active NIM endpoint to automatically populate the dropdown with all available models.
• Full OpenAI Compatibility: Instantiates standard @langchain/openai ChatOpenAI models to run natively in any advanced AI chain or agent.
• Cloud & Self-Hosted Support:
  • Point to NVIDIA Cloud catalog: https://integrate.api.nvidia.com/v1
  • Point to local/self-hosted NIM deployments: e.g., http://localhost:8000/v1
• Advanced Model Options: Customize temperature, maxTokens, topP, frequencyPenalty, presencePenalty, and timeout.

Installation

For Local Development and Testing

To test this node locally in your self-hosted n8n environment:

1. Clone/Move this directory into your development environment.
2. Run npm install to download dependencies.
3. Run npm run build to compile the TypeScript code.
4. Copy the compiled package (or link it) into your n8n custom nodes directory:
   • For standard installations, copy the folder to ~/.n8n/custom/ or run npm link in this directory, and npm link n8n-nodes-nvidia-nim inside your n8n installation directory.
   • Alternatively, you can copy the dist folder and package.json directly into your local custom nodes path.

For Production

Once published to npm (e.g. as n8n-nodes-nvidia-nim), users can install it directly via the n8n UI:

1. Go to Settings > Community Nodes.
2. Click Install a Node.
3. Enter n8n-nodes-nvidia-nim.
4. Agree to the terms and click Install.

Setup & Credentials

1. Get an API Key from the NVIDIA API Catalog or NGC portal.
2. In n8n, create a new credential for NVIDIA NIM API and input your API key.
3. Add the NVIDIA NIM Chat Model node to your workflow.
4. Connect it to the Chat Model input of an AI Agent or Model Selector node.
5. Select your desired model name (fetched dynamically from the NIM API) and start building!

License

MIT