Gemini's response cut off in half way

[TheValkyrja]

What happened?

When using Gemini endpoint and setting the model as gemini-exp-1114, AI's response cut off halfway, showing "The response is incomplete; it's either still processing, was cancelled, or censored. Refresh or try a different prompt." after refresh, the response shown as "network error". This only happens while using gemini-exp-1114 the specific model but not happening while using other geminis.

i found similar problem in #4440 , however didn't experienced the problem in gemini-1.5-pro-002 but in gemini-exp-1114.

Steps to Reproduce

1.set the model to gemini-exp-1114
2.give a relatively long promt
3.the response cut in half way.

What browsers are you seeing the problem on?

No response

Relevant log output

No response

Screenshots

Code of Conduct

• I agree to follow this project's Code of Conduct

[danny-avila]

Try updating, I don't have this issue with gemini-exp-1114

[svnkxmza]

I have the same problem. Long answers are interrupted. Short answers aren't.

[TheValkyrja]

Try updating, I don't have this issue with gemini-exp-1114

i have. this is the latest version, and i also tried re-install docker and rebuild the image, didn't work.

[kiyo-e]

Same here.

[insilications]

I have this issue with gemini-exp-1121. It is fully updated to the last docker image

[danny-avila]

At the time of moving this to discussions, it wouldn't happen to me. Now that I'm trying again, I'm also experiencing the behavior. I'm taking a closer look at this today.

[danny-avila]

Should be fixed here: #4781