Connect a self-hosted Open WebUI instance to ChinaGenAI's OpenAI-compatible Chat Completions endpoint. This setup is for short text chats; it does not add other API families to the site.
Check current availability and request limits before connecting. Check free API access and limits.
Connection values
- OpenAI API base URL
https://chinagenai.com/v1- API key
- Your ChinaGenAI site key from API access; it is separate from an OpenAI or model-provider key.
- Model ID
free, the site's automatic route label. It does not name or guarantee a particular upstream model.- Chat endpoint
https://chinagenai.com/v1/chat/completions
On 20 October 2026, a public GET https://chinagenai.com/v1/models returned the model-list entry free. That checks model discovery, not a chat completion.
Add the connection in Open WebUI
- Open Admin Settings → Connections and find the OpenAI API connections. Add a connection.
- Enter the base URL above and your ChinaGenAI site key, then save. Use the base URL ending in
/v1; do not paste the full/chat/completionsendpoint into this field. - Use Verify Connection to check the model list. When the site pool is available,
freeshould appear in the model selector. Select it in a new chat when you are ready to make a request.
Open WebUI's official OpenAI-compatible connection guide
What connection verification proves
Open WebUI's verification calls the provider's /models endpoint with a Bearer token. A successful model-list check does not send a chat prompt or prove inference is working. If the model is missing, first check the exact base URL and the current site status in API access.
For other compatible providers, Open WebUI documents that verification can fail when a provider has no /models endpoint, even though chat completions may work; those connections can require manually adding the provider's exact model ID. ChinaGenAI's model-list route returned free in the dated check above.
Check the task size before chatting
ChinaGenAI exposes Chat Completions, not the Responses API. The current shared route accepts up to 6 messages and 6,000 serialized input characters, and caps output at 256 tokens. It allows one active request per key and two across the site; daily limits are shared by network and across the site. Read current access limits before using a self-hosted chat history, which may send more context than the newest message suggests.
This endpoint does not provide embeddings, image or audio APIs. A model-list entry does not establish support for every Open WebUI feature or a particular model capability. For a short Chat Completions capture, the local response inspector checks syntax and response fields without making another model request.
ChinaGenAI is an independent project and is not affiliated with Open WebUI. These are documentation-based setup steps; this guide does not claim a live Open WebUI chat was tested.
Free API access and limits
Check the site free API access and limits before trying to obtain a site key. Read the current status and request limits before using a site key. It is not a source of official provider keys.
ChinaGenAI is not affiliated with the providers named here. This release contains setup guidance and explicitly labeled fictional examples, not claimed live model benchmarks.