Skip to main content
Cherry Studio is a desktop AI client for Windows, macOS, and Linux. It supports OpenAI-compatible custom providers, so you can connect Cherry Studio to the ScitiX Model Inference API and use ScitiX models from the desktop app. Supported models include glm-5.2 and DeepSeek-V4-Flash. For model pricing, see the model catalog.

​
Configure

The steps below use Cherry Studio v1.9.x as a reference. Newer versions may differ slightly.
  1. Open Settings → Model Provider, then click + Add at the bottom of the provider list.
  2. In the add-provider dialog, set the provider Name to ScitiX and choose the OpenAI type.
  3. On the ScitiX provider page, fill in the required fields. Cherry Studio shows a live Preview below the API Host field, such as https://api.scitix.ai/model-api/v1/chat/completions. This confirms that Cherry Studio appends the remaining path automatically. Do not end the API Host with #. A trailing # forces Cherry Studio to use the address verbatim without appending /chat/completions, which breaks the request.
  4. Under Models, click + or Fetch model list, then add the model ID glm-5.2. Custom providers do not auto-populate models. Add each model ID you want to use.
  5. Click the ✓ button next to the API Key field to test connectivity. A Connection successful toast confirms that the API Key and API Host work.
  6. Turn on the toggle in the top-right corner of the ScitiX provider page. The provider’s models will not appear in the model selector until this toggle is on.
Cherry Studio — ScitiX provider configured with API Host, key, and models

​
Verify

Go back to a conversation, open the model selector at the top, and switch to a ScitiX model shown as <model> | ScitiX. Send a message. A normal reply confirms that the connection works. Cherry Studio — chatting with a ScitiX model

​
FAQ

  • Models missing from the selector: confirm that the provider toggle in the top-right corner of the provider page is on.
  • 404 on requests: check that the API Host does not end with #. For normal use, enter https://api.scitix.ai/model-api/v1.
  • Cannot fetch the model list: custom providers require adding model IDs manually. Use + under Models and enter the ID, such as glm-5.2.
  • Assistants / translation features: set the corresponding default model to a ScitiX model in Settings → Default Model so those features also use ScitiX.
  • Choose a model with native tool calling if you rely on tool/function features.