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

LobeChat

Connect LobeChat in LobeHub to RunAPI and verify a chat.

View Markdown

LobeChat is currently distributed as LobeHub. Connect it to RunAPI through the OpenAI provider with a custom Base URL. Create a standard RunAPI API key in the Authentication Guide before continuing.

Open LobeHub

Use the official LobeHub Web or desktop application, then open its application settings.

Add the RunAPI connection

  1. Open Settings -> Language Model -> OpenAI.
  2. Enter your RunAPI API key and set the custom Base URL to https://runapi.ai/v1.
  3. Enable the OpenAI provider and save the settings.
  4. Enable an exact RunAPI model from the list. If it is absent, add a custom model with the identifier from the Model Catalog.

This is the OpenAI provider, so its Base URL includes /v1. Do not copy that suffix to a provider that uses a different public protocol.

The automatic model list and custom models are separate configuration paths. A saved custom model remains the reliable fallback when automatic discovery does not show the identifier you need.

Select the Chat Completions path

Leave Use Responses API Specification disabled for this baseline setup so the model uses OpenAI-compatible Chat Completions. Enable that option only when you have separately verified the selected model and workflow against the Responses API.

Verify the connection

Create a new chat or Agent, select the configured RunAPI model, and send Reply with the word connected. A normal complete or streamed reply confirms the active conversation can use RunAPI.

Troubleshoot

  • The reply is empty or the request returns 404: confirm the Base URL includes /v1 exactly once.
  • The model does not appear: confirm the OpenAI provider is enabled, then add the exact identifier as a custom model.
  • Vision or Skill calling fails: application capability switches do not change the selected model’s actual capabilities; check the Model Catalog.
  • Chat Completions and Responses behave differently: confirm Use Responses API Specification is disabled for this guide’s baseline path.

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