TypingMind and the Wild West API: why it does not connect yet
TypingMind lets you add custom OpenAI-compatible models, but it sends those requests straight from your browser. The Wild West API does not answer CORS preflight today, so a TypingMind custom model pointing at it fails with a network error.
Set it up
- If you want to try it anyway, or check again later: open Models in the left sidebar and click Add Custom Model.
- Set Endpoint to https://wildwestapi.com/v1/chat/completions and Model ID to outlaw-1.
- Add a custom header named Authorization with the value Bearer sk-ww-your-key.
- Set Context Length, then click Test. Today the test fails with a network or CORS error.
- For a working setup now, use SillyTavern or one of the apps listed below.
Does it work with the Wild West API?
Not yet. TypingMind's own docs for custom and local models say "Typing Mind will send requests to the local model directly from the browser" and tell you to make the server accessible from the browser. Before a cross-site request that carries an Authorization header, the browser sends an OPTIONS preflight and waits for permission. The Wild West API does not answer that check today, so the browser blocks the request before it is sent.
TypingMind's proxy setting does not change this. It swaps the endpoint for OpenAI, Claude or Gemini but the browser still makes the call. The macOS app also runs the TypingMind web app, and its docs list it as a browser origin for CORS purposes.
What to use instead
- SillyTavern: the most complete roleplay front end, and it calls the API from its own local server.
- Chatbox desktop, Jan or Cherry Studio if you want a TypingMind-style general chat app.
- Open WebUI or LibreChat if you want a self-hosted web interface you can open from any device.
If TypingMind works later
When the API answers preflight, the steps above will work as written. For long roleplay, set Context Length to what you are willing to pay for each turn (the models accept 512K tokens or more; see context window) and set a reply cap with max tokens.
Troubleshooting TypingMind
- 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 the expected result today. The request is blocked by the browser before it reaches the API, so no credit is used.
FAQ
Is my key wrong if the TypingMind test fails?
Probably not. A blocked preflight looks like a network error, not a 401. Test the same key in SillyTavern to confirm it works.
Does the TypingMind Mac app avoid CORS?
No evidence says so. TypingMind's docs treat the Mac app as a browser origin for CORS, so expect the same failure.
Can I put my own proxy in front of the API?
A proxy that adds CORS headers would work technically, but it means running and securing a server that holds your key. SillyTavern is simpler.