After downloading and installing Cherry Studio, open the model service settings interface. First, create GeekAI as a model provider (note that the provider type should be OpenAI, as all of GeekAI’s model APIs are compatible with the OpenAI interface format, including Claude and Gemini):

Then, on the model provider page, fill in the GeekAI model proxy Base URL and API_KEY:

Note: Do not include a trailing ”/” after the API address. Enter https://geekai.dev/api to ensure the complete API address is https://geekai.dev/api/v1/chat/completions.

After completing the above configuration, click the manage button to enter the model management page. Here, you can see a list of all AI models supported by GeekAI. Select the models you want to use in Cherry Studio and click the ”+” button on the right to add them:

After adding the AI models, close the model management pop-up window, and you will see that the GeekAI model provider page now displays all the models you just added. Click the check button to verify the availability of any model:

A successful connection indicates that the model is available.

Then, you can go to the default model configuration interface to configure the default models for different scenarios in Cherry Studio using the AI models provided by GeekAI:

After completing the above configuration, you can enter the Cherry Studio chat interface to converse using the AI models proxied by GeekAI:

For more information on using Cherry Studio, please refer to the official documentation. You can view the usage costs for all models on the GeekAI billing page: