Wild West API

Use n8n with an uncensored LLM API

n8n runs on a server, so its nodes can call the Wild West API without browser limits. The Base URL is a field on the OpenAI credential, and the OpenAI Chat Model node then uses that credential.

Works with Wild West API.

Set it up

  1. Sign in at wildwestapi.com/signup/ with the six-digit code sent to your email, open the dashboard and create an API key. It starts with sk-ww- and is shown in full once, so copy it somewhere safe.
  2. In n8n, open Credentials and create a new OpenAI credential.
  3. Paste your sk-ww- key into API Key and leave Organization ID blank.
  4. Set Base URL to https://wildwestapi.com/v1. The default is the OpenAI address, so this field must be changed.
  5. Save. n8n tests the credential by requesting /models.
  6. Add an AI Agent or a Basic LLM Chain, attach the OpenAI Chat Model sub-node, choose the credential, and set the model to outlaw-1.
  7. Make sure Use Responses API is off in the node.

Where each setting lives

In n8n's source the OpenAI credential has a field named url labelled Base URL, described as an override for the default base URL. The credential sends the key as a bearer token and its test calls /models on that base. The node docs say the OpenAI Chat Model node defaults to the Chat Completions API unless you turn on the Responses API, which is the path Wild West documents. Leave that option off.

Model field and options

The node lists models by asking the endpoint, but if the dropdown does not show outlaw-1, type the id into the model field in expression mode instead. The node's options include Sampling Temperature, Maximum Number of Tokens, Timeout and Max Retries. For long scenes or reports raise the token cap, and keep retries low so a failing node does not repeat an expensive call. Other ids are on the models page.

Tool calling in agents

An AI Agent node passes your connected tools to the chat model as function definitions. Most Wild West models support tool calling, so agent workflows work, but test with a small tool first. Check the model you pick on the models page.

One limit to know

Use the Chat Model sub-node, not the standalone OpenAI action node. A long-standing n8n issue asked for a custom base URL on that action node, and I could not confirm that it reads the credential's Base URL on every version. Also, the ordinary workflow runs on your server, so the no-CORS rule does not affect it. A custom web page that calls the API from a visitor's browser would.

FAQ

Does it work on n8n Cloud?

It should, since the call is made from n8n's servers and the credential has a Base URL field. I could only check the credential source, not a Cloud instance.

The credential test fails. What now?

The test requests /models under the Base URL. Check the URL ends in /v1, the key starts with sk-ww- and has no spaces, and the Organization ID is empty.

Related

Use LangChain with an uncensored LLM APILangChain's ChatOpenAI class accepts a base_url, so it can drive any OpenAI-compatible server. That gives you chains, agents and LangGraph on top of an uncensored model, with no change to how you write the rest of the app. Everything below runs from a server or a script. Use the OpenAI Node SDK with an uncensored LLM APIThe openai package for Node and TypeScript accepts a baseURL option, so pointing it at Wild West is a two-field change. Run it on a server or in a script, not in browser code. Use LiteLLM with an uncensored LLM APILiteLLM routes by model prefix. For an OpenAI-compatible server like Wild West the prefix is openai/, and the server address goes in api_base.

Uncensored AI models on one key

OpenAI and Anthropic compatible, pay as you go. New to it? Start with uncensored AI, explained.

Get an API keyBrowse models