n8n-nodes-nvidia

n8n community node for NVIDIA AI models via NVIDIA's OpenAI-compatible API.

Features

• Use NVIDIA-hosted AI models (default: moonshotai/kimi-k2.5) in n8n AI workflows
• Compatible with n8n's AI Agent, LLM Chain, and other AI root nodes
• Supports streaming, tool calling, and all standard chat model parameters
• Configurable temperature, topP, max tokens, penalties, and timeout

Installation

npm install n8n-nodes-nvidia

Or install via n8n's Community Nodes UI:

1. Open n8n → Settings → Community Nodes
2. Enter n8n-nodes-nvidia and click Install

Setup

1. Get your API key from NVIDIA Build
2. In n8n, go to Settings → Credentials → Add Credential
3. Select "NVIDIA API"
4. Enter your API key and the base URL (default: https://integrate.api.nvidia.com/v1)

Usage

1. Add an AI Agent (or other AI root node) to your workflow
2. In the "Chat Model" dropdown, select "NVIDIA Chat Model"
3. Configure your model parameters (model ID, temperature, etc.)
4. Run your workflow

Available Models

Any model available on NVIDIA Build can be used. Default is moonshotai/kimi-k2.5.

License

MIT