🚩 Report
Can't answer asks
13 hours later, it's only kind of working for me. It will usually work eventually, but it takes a minute to start working and even then it keeps failing halfway through.
What do you mean by failing halfway through
@iNeverLearnedHowToRead
?
Are you again not getting the response or getting a wrong response ?
@shivi I mean that it'll start to generate a response, then run into an error and go back to blank. It hasn't happened for a couple hours now.
Prompts are getting responses reliably again, but it's just really slow. 30-60 seconds before response starts.
@shivi Now it's just timing out over and over.
very slow in api interface. and now not working and just time out. !!!
Bruh the hell is going on I literally connot get a response back this has been going on for a whole damn day now like help🗿😭🙏
hi guys, sorry about the inconvenience. will get this resolved asap. have pinged HF staff to help out with this.
@shivi Hello dear friend. I appreciate your efforts and those of your team. However, I purchased a PRO subscription from HF to access and use the CR+ Interface API, which I had recently seen in your model. After purchasing, I realized that the model inference in this API is very slow and unstable. Now it is completely inaccessible. I hope you can overcome this challenge and improve the quality of your services.
It seems to be connected. However, please check the stability of its status and the inference speed on the API interface.
Hi guys, in the mean time, you can also access Command-R models using our HF Space. We've recently upgraded it and the UI experience is same as HF Chat.
Hello friend, the HF Space seems to not work anymore, it's giving me a white page, I can provide screenshots if you desire, but I don't know if it's on my side or just a glitch, just wanna say something and I'll wait.
For me, everything seems back to normal and working again. Thank you @shivi
It's back to normal for me too, thank you, good day or night and god bless.
Hey @Ahmet98983 , since this report is resolved, it would be great if you could close this report. Thanks!
Oh, I'm sorry, i'm quite new to this but...can i say here that the model has been generating text too slowly on the https://huggingface.co/chat/ ? Because it has been slow ever since the newest updates, however, i don't know if it's on my end or something.
"403 You don't have access to this conversation. If someone gave you this link, ask them to use the 'share' feature instead."
how to fix this error???
Hey @Ratchetrex , would recommend switching to our HF space as an alternative when Hugging Chat is down.
It was overloaded a minute ago, then it said the prompt was too long, now it's just saying no output generated.
And now it's good again. Thank you for listening to these updates.
Inference API is extremely slow or completely unresponsive.
@shivi I have several apps in production using the inference API and they are almost unusable. A simple response takes minutes.
Even trying here I cannot generate anything, I get the error message “The model is overloaded.”
Anything that can be done about the inference API? This is a real issue right now.
Inference API is extremely slow or completely unresponsive.
Same here. It usually responds for me after a couple minutes of waiting (which is crazy long) but I've started getting 502, 503, and 504 errors more frequently as well.
Man, the generation time has been really slow these past days, a few months ago this was really quick from what I recall, maybe it's a problem on my side.