Cherry Studio is an open-source AI desktop client. You can connect it to Model Studio using Token Plan (Team Edition), Coding Plan, or Pay-as-you-go.
Install Cherry Studio
Download and install Cherry Studio from the Cherry Studio download page.
Configure credentials
Open Cherry Studio, click the settings button in the upper-right corner, and then click Add in the Model Service section. Enter a provider name (for example, Token Plan (Team Edition)) and select OpenAI as the provider type.
Model Studio offers three billing plans. Choose based on your needs:
-
Token Plan (Team Edition): Seat-based subscription that deducts Credits based on token consumption.
-
Coding Plan: Fixed monthly subscription billed by the number of model calls.
-
Pay-as-you-go: Postpaid billing based on actual usage.
Token Plan (Team Edition)
|
Parameter |
Description |
|
API key |
Token Plan (Team Edition) dedicated . |
|
API address |
|
|
Model |
Token Plan (Team Edition) Overview |
Coding Plan
|
Parameter |
Description |
|
API key |
Coding Plan dedicated API key. |
|
API address |
|
|
Model |
Coding Plan Overview |
Pay-as-you-go
|
Parameter |
Description |
|
API key |
See Obtain an API key. |
|
API address |
Enter the URL for your region:
|
|
Model |
See Supported models. |
Verify configuration
In the Model ID field, enter the ID of the model you want to use, such as qwen3.7-max, and then click Add. Return to the chat interface and ask a question. A valid response confirms the setup.
If you are a RAM user, see Manage workspaces to check model access permissions.
FAQ
Error codes
If you encounter errors, see the FAQ for your billing plan:
-
Pay-as-you-go: Error codes
-
Coding Plan: FAQ
-
Token Plan (Team Edition): FAQ
Error: The value of the enable_thinking parameter is restricted to True
Cause: This model requires thinking mode, which was not enabled.
Solution: Enable thinking mode in the client.
Charges incurred despite having a free quota when using Pay-as-you-go
Possible causes:
-
Region mismatch: The free quota applies only to models in the Singapore region (international deployment scope). Using models in other regions or deployment scopes incurs charges. Ensure the API address matches your target region. See Free quota for new users.
-
Per-model free quota: Each model has its own free quota.
-
Data update delay: Free quota data updates hourly. Your quota may be exhausted even if the console shows a remaining balance.
See Identify models with charges and How do I view model invocation records? for details.