Add an uncensored custom provider to Cherry Studio
Cherry Studio is an open-source desktop chat client for Windows, macOS and Linux. Custom providers of type OpenAI work with any compatible API, and the desktop app is not subject to browser CORS rules, so the Wild West API works.
Set it up
- Sign in at wildwestapi.com/signup/ with the six-digit code sent to your email, open the dashboard and create an API key. It starts with sk-ww- and is shown in full once, so copy it somewhere safe.
- Click the gear icon in the left bar and open the Model Services tab.
- Click Add Provider, enter Wild West as Provider Name and choose OpenAI as Provider Type.
- Paste your sk-ww- key into API Key.
- Set API Address to https://wildwestapi.com (Cherry Studio adds /v1 itself if it is missing).
- Click Get model list, or use the plus button to add outlaw-1 by hand.
- Switch the provider on, then pick outlaw-1 in a new topic.
Does it work with the Wild West API?
Yes. Cherry Studio is an Electron app whose main window runs with web security turned off, which removes the browser's cross-site check. Requests reach the API even though it does not answer CORS preflight.
About the API Address
Cherry Studio completes the address for you. In current versions it appends /v1 unless the address already has a version segment, then adds /chat/completions. Both of these work:
https://wildwestapi.com https://wildwestapi.com/v1
Ending the address with # tells Cherry Studio to use it exactly as typed, which is only needed for unusual paths.
The settings that matter for long roleplay
- In the assistant's model settings, set the number of context messages. A low count is why characters forget earlier scenes. These models accept 512K tokens or more of context window, but more history costs more each turn.
- Set max tokens high enough for full replies; see max tokens.
- Keep streaming output on; see streaming.
- Put your character in the assistant prompt; see system prompt.
Troubleshooting Cherry Studio
- 401: the key is missing, mistyped or disabled. Paste it again with no trailing space and check it starts with
sk-ww-. A key from another provider will never work here. - 402: your balance is too low for the request. Top up in the dashboard; see pricing. A huge context on every turn burns credit faster than you might expect.
- 404 or 400 with an unknown model: the model id is wrong. Use an exact id from the model list, such as
outlaw-1, with no provider prefix and no spaces. - 429: you are rate limited. Wait a few seconds and send again, and turn off any auto-retry loop that fires instantly.
- 5xx: trouble upstream. Retry; if it keeps happening, switch to another model id for a while.
- Network error, "Failed to fetch" or a CORS message: this is unusual in Cherry Studio. Check the API Address for typos and that the provider is switched on.
- 404 with /v1/v1 in the log: older Cherry Studio versions handled the trailing slash differently. Try
https://wildwestapi.comwith no path, or update the app.
FAQ
Which Provider Type should I choose?
OpenAI. That makes Cherry Studio use the chat completions format the API serves.
Can I add several models at once?
Yes. Get model list reads /v1/models and lets you add the ones you want.
Does Cherry Studio need my key on every model?
No. The key belongs to the provider, and every model under it uses that key.