boldrouter for n8n

You can now use boldrouter as the language model in n8n workflows. The BoldRouter Chat Model node connects to the Model input of an AI Agent, a Basic LLM Chain, a Question and Answer Chain, a Summarization Chain, or any other n8n node that takes a chat model.

Install

On a self-hosted n8n, open Settings → Community Nodes → Install and enter:

n8n-nodes-boldrouter

The node needs n8n 2.x.

Connect it

1. Create an API key on the API keys page. Keys start with sk-bold-.
2. In n8n, add a BoldRouter API credential and paste the key. n8n checks it straight away.
3. Add an AI Agent (or a chain), click its Model connector and choose BoldRouter Chat Model.
4. Pick a model. The list is loaded live from your account, so it shows exactly the models you can use. bold/auto is the default and lets boldrouter choose the model for each request.

What you can set

Under Options, the node exposes the same controls as the API:

• Routing: a strategy (balanced, price, latency or uptime), a preferred provider order, providers to use only or to ignore, and whether to fail over.
• Fallback models: other models to try in order if the first one fails, for example openai/gpt-5.5, google/gemini-2.5-pro.
• Reasoning effort for models that think before answering.
• Anonymize prompts: personal data is removed before the prompt reaches the provider and put back in the answer.
• Prompt caching on or off, plus the usual temperature, top P, max tokens, penalties, timeout and retries.

Streaming, tool calls and images work as they do with any n8n chat model, so an AI Agent's tools run unchanged.

The node is open source on GitHub and published on npm. See Authentication and Routing for what each option does.