During the use of the Gemini API, there have been a large number of empty replies and truncation phenomena, which affect the normal use of my program, plz fix it.This really affects my work
i think they dont even know that its broken alot of users have been reporting this and its been happening for like 3 days
yes, i suspect some TPUs are being used for training new models, but the normal use of old models should also be guaranteed.
i dont think its a tpu thing tbh i think they are getting the api ready for the new model maybe ? or maybe they didnt notice that there is a problem
Thanks for bringing this up ,
The engineering team is aware of the empty response issue and is actively working on a fix. We will keep you updated as the fix is identified and rolled out.
Thanks for updating on the progress. It’s very troublesome. It’s been there for a long time.
Can you give us a rough idea how long will it take to get fixed?
Our workflow it pretty much devastated by the issue.