Run the EveryAPI gateway inside n8n
n8n is a workflow automation platform with visual, code, and self-hosted deployment options. Where an OpenAI credential or model node exposes a custom Base URL, point that specific credential at EveryAPI and select a model available to your account.
One base URL line points n8n at EveryAPI
Default global Base URL / China-first Base URL: https://api.everyapi.ai/v1 / https://api-cn.everyapi.ai/v1
# 1. In n8n, Credentials → New Credential → search "OpenAI"# 2. Fill in two fields API Key sk-everyapi-...Base URL https://api.everyapi.ai/v1 # No /chat/completions path segment # 3. Save. From now on, every OpenAI / LangChain OpenAI Chat Model node can pick this credentialFrom zero to your first call
- Deploy n8n
Use n8n Cloud or a supported self-hosted release. Credential fields and AI nodes can vary by n8n version.
n8n.io/docs/hosting
- Create an OpenAI Credential
Create an OpenAI credential only if your n8n version exposes a custom Base URL. Set it to https://api.everyapi.ai/v1 and store the EveryAPI key in the credential.
- Select the credential in workflow nodes
Attach that credential to a compatible OpenAI or chat-model node and enter a model ID available in your EveryAPI account. Verify that the node permits a custom model ID.
- Test the workflow behavior
Run a small workflow first. Tool calling, structured output, memory, and streaming depend on the n8n node and selected model; an OpenAI-compatible endpoint does not guarantee identical behavior.
n8n × EveryAPI
Why doesn't the Base URL include /chat/completions?
Internally n8n uses the standard OpenAI SDK, which builds the /v1/chat/completions path segment itself. The Base URL field just needs the root + /v1 (https://api.everyapi.ai/v1). This is consistent with ChatUIs like LobeChat.
Can n8n's Anthropic / Google nodes switch to EveryAPI?
Use an OpenAI credential or model node only when your n8n version exposes a custom Base URL for it. Enter a model ID from your EveryAPI account catalog; other provider-specific n8n nodes keep their own credential and protocol requirements.