Description

n8n-nodes-nvidia-nim

!n8n.io – Workflow Automation
!npm version
!License: MIT

n8n community node for NVIDIA NIM – Chat completions and image analysis with NVIDIA AI models.

📋 Requirements

• n8n version 1.0.0 or higher
• Node.js v18.17.0 or higher
• NVIDIA NGC API Key – Get yours at ngc.nvidia.com

📦 Installation

Via n8n Community Nodes (Recommended):

1. Go to Settings → Community Nodes
2. Click Install
3. Enter: n8n-nodes-nvidia-nim
4. Restart n8n after installation

Via npm:

npm install n8n-nodes-nvidia-nim

⚙️ Setup

1. Get NVIDIA API Key: ngc.nvidia.com
2. Add Credentials in n8n:
– Go to Credentials → New
– Select NVIDIA NIM API
– Enter API Key and Base URL: https://integrate.api.nvidia.com/v1

🎯 Basic Usage

Text Chat

1. Add “NVIDIA NIM” node → Configure model (e.g., meta/llama3-8b-instruct)
2. Connect: Trigger → NVIDIA NIM (main)
3. Execute workflow

Image Analysis

1. Add “NVIDIA NIM Image Analysis” node → Configure vision model (e.g., meta/llama-3.2-11b-vision-instruct)
2. Provide image data (URL, base64, or data URL) and analysis prompt
3. Connect: Trigger → NVIDIA NIM Image Analysis (main)
4. Execute workflow

🤖 Available Models

Text Models

Current recommended models include:

• meta/llama-3.1-8b-instruct ⭐ (recommended, fast & efficient)
• meta/llama-3.1-70b-instruct (high quality)
• meta/llama-3.1-405b-instruct (best performance)
• nvidia/nemotron-4-340b-instruct (enterprise-grade)
• mistralai/mixtral-8x7b-instruct-v0.1
• deepseek-ai/deepseek-r1 (reasoning models)
• google/gemma-2-27b-it
• snowflake/snowflake-arctic-instruct

View all models →
Try NVIDIA NIM APIs →

Vision Models

• meta/llama-3.2-11b-vision-instruct ⭐ (recommended)
• meta/llama-3.2-90b-vision-instruct

View all vision models →

⚙️ Configuration

Key Parameters:

• Model: Choose from NVIDIA models
• Temperature: 0.0-2.0 (default: 0.7)
• Max Tokens: Response length (default: 1024)
• Top P: Nucleus sampling (default: 1.0)

📚 Resources

• CHANGELOG – Version history
• GitHub Repository
• npm Package
• NVIDIA NIM Docs

🤝 Contributing

Issues and PRs welcome on GitHub

—

Made by Akash Kumar Naik