Description
n8n-nodes-iointelligence
This is an n8n community node that lets you use IO Intelligence as an AI model provider in your n8n workflows.
IO Intelligence provides access to a range of open-source and proprietary LLMs through io.net’s decentralized GPU infrastructure.
Installation
Follow the installation guide in the n8n community nodes documentation.
npm package name: n8n-nodes-iointelligence
Credentials
You need an IO Intelligence API key to use this node:
1. Go to io.net and create an account
2. Navigate to your API settings and generate an API key
3. In n8n, create new IO Intelligence API credentials and paste your API key
The default base URL is https://api.intelligence.io.solutions/api/v1. You can change this in the credentials if needed.
Node: IO Intelligence Chat Model
This node provides a Chat Model that integrates with n8n’s AI workflow system. It appears as an AI Language Model sub-node that you can connect to:
- AI Agent node
- Basic LLM Chain node
- Other AI chain/agent nodes that accept a chat model input
- Tested with n8n version 1.x
- Requires Node.js 18+
- Self-hosted n8n: Fully supported. Install via Community Nodes in the n8n UI.
- n8n Cloud: This node uses
@langchain/openaias a peer dependency for AI Agent/Chain integration. Cloud support depends on n8n’s verification of AI community nodes. - IO Intelligence API Docs
- n8n Community Nodes Documentation
Configuration
| Parameter | Description |
|———–|————-|
| Model | Select from available models (fetched dynamically from the API) |
| Temperature | Controls randomness (0 = deterministic, 2 = very creative) |
| Maximum Tokens | Max tokens to generate (-1 for no limit) |
| Top P | Nucleus sampling parameter |
| Frequency Penalty | Penalizes repeated tokens |
| Presence Penalty | Encourages new topics |
| Timeout | Request timeout in milliseconds |
Usage
1. Add an AI Agent or Basic LLM Chain node to your workflow
2. Click the Model input connector
3. Select IO Intelligence Chat Model from the list
4. Configure your credentials and select a model
5. Run your workflow