Skip to content
RunAPI Developer Docs
EN 中文
Developer Resources
Developer Resources

NextChat

Connect NextChat to RunAPI with a custom endpoint and model.

View Markdown

Connect NextChat to RunAPI through its custom OpenAI endpoint. NextChat composes its operation path differently from the other clients in this group, so use the exact endpoint value below. Create a standard RunAPI API key in the Authentication Guide before continuing.

Install or deploy NextChat

Choose a desktop release or deployment method from the official NextChat repository, then open Settings.

Configure the custom endpoint

  1. In the Web application, enable Custom Endpoint and select OpenAI. The desktop application always uses custom configuration.
  2. Set the endpoint address to https://runapi.ai without /v1.
  3. Enter your RunAPI API key.
  4. Under Custom Model Name, add an exact identifier from the Model Catalog, then select it for the current chat.

This setting is NextChat’s endpoint root, not the OpenAI-compatible Base URL used by an OpenAI SDK. NextChat appends v1/chat/completions itself, producing https://runapi.ai/v1/chat/completions. Its OpenAI client does not fetch the RunAPI model list, so add the identifier manually instead of waiting for discovery.

Configure a self-hosted deployment

For a self-hosted installation, set these environment variables and redeploy or restart the application:

SHELL 
BASE_URL=https://runapi.ai
OPENAI_API_KEY=YOUR_RUNAPI_API_TOKEN
CUSTOM_MODELS=+YOUR_RUNAPI_MODEL_ID

The + prefix adds the model without replacing the application’s existing model entries.

Verify the connection

Start a new chat, select the custom RunAPI model, and send Reply with the word connected. A normal complete or streamed response confirms the custom endpoint is active.

Troubleshoot

  • The request URL contains /v1/v1/chat/completions: remove /v1 from the endpoint address and use https://runapi.ai.
  • NextChat still uses a built-in model: add the exact identifier under Custom Model Name, then reselect it in the current chat.
  • Web settings have no effect: confirm Custom Endpoint is enabled; the desktop and Web applications handle this switch differently.
  • Self-hosted changes have no effect: restart or redeploy after changing BASE_URL, OPENAI_API_KEY, or CUSTOM_MODELS.

For shared protocol behavior, use the LLM API Quickstart. For request and response fields, use the Chat Completions API Reference.