GPT OSS models are running very slow from Yesterday?
Earlier just simple greeting message used to take 0.7 ms and now it takes 5 seconds or more with API as well as in playground.
Did something happened? However other models are working fast. See the attached screenshot
@ritesh.khapre
Thank you for sharing this feedback. We appreciate you bringing this to our attention.We will check this issue on our side and get back to you with an update shortly.
it looks like this got resolved, may we know the reason or it was from our end only?
@ritesh.khapre it did not impact only you. It was an internal bug that got fixed.
-Coby
similar issue now we are facing for maverick model as well.
- SCX here is our system running on SN40