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

Cherry Studio

Connect Cherry Studio to RunAPI and verify a chat.

View Markdown

Connect Cherry Studio to RunAPI with a custom OpenAI provider. Create a standard RunAPI API key in the Authentication Guide before continuing.

Install Cherry Studio

Install the desktop application by following the official Cherry Studio download guide, then open its settings.

Add the RunAPI provider

  1. Open Settings -> Model Services -> + Add.
  2. Set the provider name to RunAPI and Provider type to OpenAI.
  3. Enter your RunAPI API key and set the API address to https://runapi.ai/v1.
  4. Add an exact model identifier from the Model Catalog, enable the provider, and save.

Because the provider type is OpenAI, this field uses RunAPI’s OpenAI-compatible Base URL, which includes /v1. Cherry Studio adds /chat/completions for a Chat Completions request, so do not paste the complete operation URL into the API address field.

Select a model

Custom providers rely on the models you add. If the selector is empty, return to the provider’s model management area and add an exact RunAPI identifier manually. Adding a model does not grant capabilities that the model does not support.

Run the connection check

Run Connection check with the saved model. This sends a model request rather than only validating the key format. A successful check confirms the provider URL, key, and selected model work together.

Verify the connection

Start a new chat, select the RunAPI provider and model, then send Reply with the word connected. A normal complete or streamed reply confirms the active chat is using RunAPI.

Troubleshoot

  • The model selector is empty: add the exact model identifier in the custom provider’s model management area and confirm the provider is enabled.
  • Connection check fails: confirm a model is selected, replace the key with a valid standard key, and retry.
  • The request returns 404: use https://runapi.ai/v1 as the API address; do not append /chat/completions yourself.
  • Text works but vision or tools fail: choose a model that declares the required capability in the Model Catalog.

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