Skip to main content
n8n is a workflow automation platform. The Eden AI Chat Model node plugs Eden AI into any AI Agent or Basic LLM Chain, so one credential covers every chat model in the catalog.

​
Overview

The node comes from Eden AI’s own community package, n8n-nodes-edenai. Use it as the language model for any n8n AI node, with no direct API wiring.
  • 500+ chat models: OpenAI, Anthropic, Google, Mistral and more through a single credential
  • Automatic fallback: define backup models that are tried if the primary fails
  • Web search: give models real-time web access when the task needs it
  • EU Only switch: route through the EU endpoint and list only EU-eligible models
The package also ships an Eden AI Embeddings node and an Eden AI Expert Models node (see Other nodes in the package).

​
Prerequisites

  • n8n 2.16 or newer. The nodes are built on @n8n/ai-node-sdk, which older n8n releases lack.
  • An Eden AI API key from app.edenai.run

​
Setup

​
1. Install the community node

In your n8n instance, go to Settings > Community Nodes and install:
Restart n8n if prompted.

​
2. Add your Eden AI credential

Create a credential of type Eden AI API: search for it when you add a credential, or create it inline from the node’s Credential dropdown. Paste your key into the API Key field. The base URL is preset to https://api.edenai.run/v3, so there is nothing else to fill in. When you save, n8n tests the credential with a request to /models, so an invalid key fails at that point.

​
3. Add the node to a workflow

In any workflow that uses an AI Agent or Basic LLM Chain node, click the Model sub-node slot and select Eden AI Chat Model. The node connects as a language model provider, and n8n handles message formatting, memory and tool calls.

​
Configuration

These fields are on the node itself: These fields are under Options > Add Option: Fallbacks and Smart Routing are Eden AI features: see Provider routing. Web search is covered in Web search and the response formats in Structured output. With Response Format set to JSON, the node’s own notice asks you to include the word json in the prompt.

​
Model naming

Models use the provider/model format, for example openai/gpt-latest, anthropic/claude-sonnet-latest or vertex/gemini-flash-latest. The Model dropdown lists the live chat catalog, so you can also pick Smart Routing (@edenai). See List LLM Models for the catalog.

​
Other nodes in the package

  • Eden AI Embeddings is an AI sub-node for vector stores. Its Model dropdown lists the embeddings catalog (see Embeddings), and it has the same EU Only switch.
  • Eden AI Expert Models is an action node that runs the non-chat features, such as OCR, text analysis, image, audio and translation. See Expert models.

​
EU data residency

Switch on EU Only in the Eden AI Chat Model or Eden AI Embeddings node. The model list then shows only models served on api.eu.edenai.run, which answers 451 for any other model. See EU Data Residency Endpoint.

​
Troubleshooting

​
The credential test fails

n8n tests the credential with a request to /models. Check that the key is an Eden AI API key and that you pasted it into API Key.

​
Next Steps