[GH-ISSUE #1348] docs: README OpenAI-only install omits OPENAI_API_BASE_URL #163

New issue

Closed

opened 2026-10-18 14:09:33 +00:00 by bsakaguchi · 1 comment

bsakaguchi commented

2026-10-18 14:09:33 +00:00

Owner

Originally created by @seven7763 on GitHub (Aug 2, 2026).
Original GitHub issue: https://github.com/open-webui/docs/issues/1348

Problem

The README section Installation for OpenAI API Usage Only only documents OPENAI_API_KEY.

The backend already supports OPENAI_API_BASE_URL / OPENAI_API_BASE_URLS (see backend/open_webui/config.py) for any OpenAI-compatible endpoint, and Admin → Settings → Connections can configure base URLs after startup. New users wiring a compatible gateway or local OpenAI-spec server often miss this.

Proposal

Add a short README example showing:

-e OPENAI_API_BASE_URL=https://api.example.com/v1
-e OPENAI_API_KEY=...

plus a note that Admin → Connections works post-startup, and to use a model ID the target actually serves.

Happy to open a docs PR against dev with CLA if this is wanted.

Originally created by @seven7763 on GitHub (Aug 2, 2026). Original GitHub issue: https://github.com/open-webui/docs/issues/1348 ## Problem The README section **Installation for OpenAI API Usage Only** only documents `OPENAI_API_KEY`. The backend already supports `OPENAI_API_BASE_URL` / `OPENAI_API_BASE_URLS` (see `backend/open_webui/config.py`) for any OpenAI-compatible endpoint, and Admin → Settings → Connections can configure base URLs after startup. New users wiring a compatible gateway or local OpenAI-spec server often miss this. ## Proposal Add a short README example showing: ```bash -e OPENAI_API_BASE_URL=https://api.example.com/v1 -e OPENAI_API_KEY=... ``` plus a note that Admin → Connections works post-startup, and to use a model ID the target actually serves. Happy to open a docs PR against `dev` with CLA if this is wanted.

bsakaguchi closed this issue

2026-10-18 14:09:33 +00:00

bsakaguchi commented

2026-10-18 14:09:33 +00:00

Author

Owner

@owui-terminator[bot] commented on GitHub (Aug 2, 2026):

🔍 Related Issues Found

I found some existing issues that might be related. Please check if any of these are duplicates or contain helpful solutions:

🟣 #22322 issue: OpenAI API Connection Model-IDs are not saved
This issue is directly about OpenAI API connection base configuration in the Admin panel, specifically model IDs not persisting after refresh. It is closely related to documenting connection setup and model IDs for OpenAI-compatible endpoints.
by m1g32 · bug
🟣 #26464 issue: Can't create a new connection (to LiteLLM)
The reporter mentions LiteLLM/OpenAI-style connection setup and confusion about the OpenAI endpoint. While not about README docs, it highlights users trying to add non-OpenAI base URLs and struggling with connection configuration.
by hpcpony · bug
🟣 #14717 issue: Cannot use OpenAI-compatible server API in Playground Completions mode
This issue shows that OpenAI-compatible server URLs were not being used in one UI flow, which is adjacent to the new issue’s concern that users need clearer documentation for base URL configuration.
by secminhr · bug
🟣 #14543 issue: Custom models based on those from OpenAI compatible API are not visible in the models list
This issue involves OpenAI-compatible API models not appearing correctly in the UI after being configured. It is relevant because the new docs request also stresses that the target server must actually serve the model ID.
by cybergod96 · bug

💡 If your issue is a duplicate, please close it and add any additional details to the existing issue instead.

This comment was generated automatically. React with 👍 if helpful, 👎 if not.

@owui-terminator[bot] commented on GitHub (Aug 2, 2026):  🔍 **Related Issues Found** I found some existing issues that might be related. Please check if any of these are duplicates or contain helpful solutions: 1. 🟣 [#22322](https://github.com/open-webui/open-webui/issues/22322) **issue: OpenAI API Connection Model-IDs are not saved** *This issue is directly about OpenAI API connection base configuration in the Admin panel, specifically model IDs not persisting after refresh. It is closely related to documenting connection setup and model IDs for OpenAI-compatible endpoints.* *by m1g32 · `bug`* 2. 🟣 [#26464](https://github.com/open-webui/open-webui/issues/26464) **issue: Can't create a new connection (to LiteLLM)** *The reporter mentions LiteLLM/OpenAI-style connection setup and confusion about the OpenAI endpoint. While not about README docs, it highlights users trying to add non-OpenAI base URLs and struggling with connection configuration.* *by hpcpony · `bug`* 3. 🟣 [#14717](https://github.com/open-webui/open-webui/issues/14717) **issue: Cannot use OpenAI-compatible server API in Playground Completions mode** *This issue shows that OpenAI-compatible server URLs were not being used in one UI flow, which is adjacent to the new issue’s concern that users need clearer documentation for base URL configuration.* *by secminhr · `bug`* 4. 🟣 [#14543](https://github.com/open-webui/open-webui/issues/14543) **issue: Custom models based on those from OpenAI compatible API are not visible in the models list** *This issue involves OpenAI-compatible API models not appearing correctly in the UI after being configured. It is relevant because the new docs request also stresses that the target server must actually serve the model ID.* *by cybergod96 · `bug`* --- 💡 If your issue is a duplicate, please close it and add any additional details to the existing issue instead. *This comment was generated automatically.* React with 👍 if helpful, 👎 if not.

bsakaguchi referenced this issue

2026-10-18 14:10:53 +00:00

[PR #163] [MERGED] Fix TTS environment variable documentation #293