Originally created by @SimplyCorbett on GitHub (Dec 18, 2024).
I am running Open-WebUI in Fedora 41 with a 3.11 python venv.
When using text completion in SillyTavern, it tries to connect to the very first (top) server in OpenAI connections.
If that server is disabled or otherwise unavailable, the connection will fail.
I have tried to change the model manually, but it doesn't change anything.
In comparison, Chat Completion will work flawlessly and allow you to select the model which routes to the right server.
Originally created by @SimplyCorbett on GitHub (Dec 18, 2024).
I am running Open-WebUI in Fedora 41 with a 3.11 python venv.
When using text completion in SillyTavern, it tries to connect to the very first (top) server in OpenAI connections.
If that server is disabled or otherwise unavailable, the connection will fail.
I have tried to change the model manually, but it doesn't change anything.
In comparison, Chat Completion will work flawlessly and allow you to select the model which routes to the right server.


Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @SimplyCorbett on GitHub (Dec 18, 2024).
I am running Open-WebUI in Fedora 41 with a 3.11 python venv.
When using text completion in SillyTavern, it tries to connect to the very first (top) server in OpenAI connections.
If that server is disabled or otherwise unavailable, the connection will fail.
I have tried to change the model manually, but it doesn't change anything.
In comparison, Chat Completion will work flawlessly and allow you to select the model which routes to the right server.