Chatbox
Chatbox is a cross-platform desktop chat client
(macOS, Windows, Linux). It runs natively, so it reaches the proxy on
localhost directly — no Docker networking to worry about.
Before you start
Do the shared setup from How-to guides first: the proxy must be running and funded.
Connect it
Add a custom provider
Open Settings, go to the providers list, and click Add Custom Provider.
Fill in the fields
- API Mode: OpenAI API Compatible
- Name:
ZeroSignal(anything you like) - API Host:
http://localhost:9376/v1 - API Path: leave the default (
/chat/completions) - API Key: anything non-empty — the proxy ignores it
- Model: a model id from the catalog (see
Picking a model), e.g.
qwen3-coder
Save and chat
Save the provider, select it (and your model) in a new chat, and send a message. The first reply confirms the path works end to end.
info
Chatbox asks for a Model by name rather than auto-loading the catalog. Add
the model ids you want from curl http://localhost:9376/v1/models — you can
add several to the provider and switch between them per chat.
Troubleshooting
| Symptom | Fix |
|---|---|
| Can't connect | Confirm the proxy is up: curl http://localhost:9376/healthz. Check the API Host ends in /v1 and has no trailing /chat/completions. |
| 404 on send | The base URL is off. Set API Host to exactly http://localhost:9376/v1 and leave API Path at its default. |
| Model "not found" | You typed a model id the network doesn't serve. Use one from curl http://localhost:9376/v1/models. |
wallet_unfunded / payment errors | zs-proxy fund — see Wallet & funding. |
What's next
- Pricing — what a message costs.
- Configuration — spend caps and the output ceiling.
- How-to guides — connect another app.