LobeChat → AI24X
LobeChat (LobeHub) is a popular open-source chat client. Add AI24X as a custom OpenAI-compatible provider—one key for flash/pro and VIP named models, PayPal USD top-up.
1) Web setup (tested path)
- Open lobechat.com (or the desktop app).
- Settings → Language models → Add custom provider → choose OpenAI.
- Provider id / name:
AI24X. Short blurb: China LLMs · one key · great value. - API Key: your AI24X key from Console.
- API proxy / base URL:
https://api.ai24x.com/v1— stop at/v1(do not append/chat/completions). - Use Responses API: off (required today; otherwise you get HTTP 405).
- Fetch model list → enable
flash(andpro/ultraif you need them) → Save. - New chat → pick AI24X / flash → send
hello.
2) Desktop
Windows / macOS desktop uses the same path: Settings → Language models → custom provider → same fields as above.
3) Pitfalls (field-tested)
- Proxy set to
…/v1/chat/completions→ 405 (LobeChat already appends/chat/completions). - Responses API left on → 405 (
/v1/responsesnot on the platform yet; planned). - Connectivity check may fail while chat still works (check may use GET; chat uses POST).
- After rotating a key in Console, update the provider’s API Key field.
4) Quick errors
405 → fix base URL to /v1 or turn off Responses. 401 → new key in Console. 429 → slow down and retry.