Cherry Studio BetterToken setup: custom provider, API Host, and models
Add BetterToken to Cherry Studio as a custom OpenAI-compatible or Anthropic-compatible provider, add Model IDs, check the connection, and fix common errors.
To connect Cherry Studio to BetterToken, add a custom provider, enter your BetterToken API Key and API Host, then add a current Model ID from the model plaza.
Key settings
| Field | Value |
|---|---|
| Provider type | OpenAI (OpenAI-compatible) |
| API Key | BetterToken API Key |
| Base URL | https://bettertoken.ai (API Host) |
| Model | A current Model ID from the model plaza |
Prerequisites
- The current version of Cherry Studio is installed
- A BetterToken API Key: register and get one
- A Model ID from the model plaza
Already have an account? Sign in, top up, then create a key in the console.
Install
Download Cherry Studio for your system from the official website or GitHub Releases, install it, and open it. No extra plugin is needed.
Manual setup
Complete this setup in the Cherry Studio settings UI. You do not need to edit a configuration file.
Add a custom provider
Open Settings → Model Provider, click Add, enter a name such as BetterToken, and choose the provider type OpenAI.
Enter the API Key and API Host
- API Key → your BetterToken API Key
- API Host →
https://bettertoken.ai
Cherry Studio appends the request path itself. Check the request URL preview under the field: it should read https://bettertoken.ai/v1/chat/completions. Do not paste a full /chat/completions URL into API Host.
Add models
Click Manage to fetch the model list, or click Add and enter a Model ID manually, for example claude-sonnet-5-5, gpt-6-sol, or deepseek-flash.
Check the connection
Click Check, select one of the added models, and wait for the success message. Then enable the provider.
Use the Anthropic protocol instead
For Claude models, you can add a second custom provider with the provider type Anthropic and API Host https://bettertoken.ai. Cherry Studio then sends requests to /v1/messages. Use this provider only for Claude models. Use the OpenAI provider type for all other models.
Verify the connection
Start a new conversation, choose a BetterToken model in the model selector, and send Reply with one word: OK. A normal reply means the setup works.
Switch models
Choose another BetterToken model in the model selector of the conversation. To use a model that is not in the list yet, open the BetterToken provider, click Add, and enter its exact Model ID from the model plaza.
Common errors
| Error | What to check |
|---|---|
401 or Unauthorized | Check the BetterToken API Key and remove leading or trailing spaces. |
404 | Check that API Host is https://bettertoken.ai and that the request URL preview ends with /v1/chat/completions only once. |
model not found | Copy a current Model ID from the model plaza. |
| Empty model list | Add the Model ID manually with Add. |
Advanced setup
Related guides
Technical details
Protocol and endpoint
The OpenAI provider type sends requests to https://bettertoken.ai/v1/chat/completions. The Anthropic provider type sends requests to https://bettertoken.ai/v1/messages. In both cases enter only https://bettertoken.ai as API Host.

