Connect n8n to Eden AI using the official community node, and plug +300 LLM models from 50+ providers into any AI Agent or Chain workflow.Documentation Index
Fetch the complete documentation index at: https://www.edenai.co/docs/llms.txt
Use this file to discover all available pages before exploring further.
Overview
n8n is a workflow automation platform with native AI agent support. The Eden AI Chat Model community node lets you use Eden AI as the language model backend for any n8n AI node — no direct API wiring required.- +300 LLM models: OpenAI, Anthropic, Google, Mistral, and more through a single credential
- Automatic fallback: define backup models that activate if the primary fails
- Web search: give models real-time web access when the task needs it
- European hosting: GDPR-compliant, transparent data routing
Setup
1. Install the community node
In your n8n instance, go to Settings → Community Nodes and install:2. Add your Eden AI credential
In n8n, click the + button → select Credential → search for Eden AI. Paste your API key from app.edenai.run → API Keys.3. Add the node to a workflow
In any workflow that uses an AI Agent or Chain node, click the Model sub-node slot and select Eden AI Chat Model. The node connects as a language model provider — n8n handles message formatting, memory, and tool calls automatically.Configuration
| Option | Description |
|---|---|
| Model | Model in provider/model format. Loaded dynamically from Eden AI’s catalog. |
| Fallback Models | Comma-separated backup models (e.g. openai/gpt-4o,anthropic/claude-haiku-4-5). Retried in order if the primary fails. |
| Web Search | Allow the model to retrieve real-time information from the web. |
| Web Search Context Size | low, medium, or high — controls retrieval depth and cost. |
| Sampling Temperature | Controls randomness (0–2). Lower = more deterministic. |
| Maximum Number of Tokens | Max tokens to generate. Use -1 for the model’s default maximum. |
| Response Format | text (default) or json_object (forces valid JSON output). |
| Top P | Nucleus sampling threshold (0–1). Alter this or temperature, not both. |
| Frequency Penalty | Reduces repetition of frequent tokens (-2 to 2). |
| Presence Penalty | Encourages new topics by penalizing already-used tokens (-2 to 2). |
| Max Retries | How many times to retry on failure (default: 2). |
| Timeout | Request timeout in milliseconds (default: 360 000). |
Available models
Pass anyprovider/model string as the model. The dropdown is populated live from Eden AI’s catalog. Examples:
openai/gpt-4o— GPT-4oanthropic/claude-sonnet-4-5— Claude Sonnetgoogle/gemini-2.5-flash— Gemini Flashmistral/mistral-large-latest— Mistral Large
JSON response format
When Response Format is set tojson_object, you must include the word json somewhere in the prompt. This is a requirement of the underlying model API.
JSON mode is only supported by models released after November 2023.
Next Steps
- LangChain - Build LLM applications in Python or TypeScript
- OpenAI SDK (Python) - Direct API access with the OpenAI-compatible SDK
- OpenAI SDK (TypeScript) - Direct API access in TypeScript