Lobe Chat

Open-source web chat framework (self-hostable). Point the OpenAI provider's proxy URL at the site.

Prerequisites

Signed in with balance · key created in the right group (create a key) · model IDs from the models table or /v1/models

Download & install

Hosted at lobehub.com, or self-host via Docker / Vercel (repo lobehub/lobe-chat).

Setup steps

1
Settings → AI Provider → OpenAI.
2
API Key = sk-xxxx; API proxy URL = https://api.daoxe.com/v1 (the /v1 suffix is required).
3
Enable client-side request mode so calls go straight from the browser instead of Lobe's server.
4
Fetch the model list automatically or add model IDs manually, then run the connectivity check.

Choosing models

Model IDs must match the site exactly (case-sensitive). Copy them from the recommended models table or GET /v1/models. If the key's group doesn't include the model you'll get 403 / model not available — see Models & groups.

Common pitfalls

Forgetting /v1 in the proxy URL is the most common mistake — symptoms are 404 / Connection failed.
The hosted version may fail to reach custom endpoints through its server relay; client-side request mode fixes this.

UI reference

Lobe Chat OpenAI provider settings — Lobe Chat · OpenAI provider settings · Click image to enlarge