Open hanmindev opened 11 months ago
I just tried running a couple generations (also was trying that throughout the day), but didn't get any errors as shown above. Did you always get an error when you run the model? As stated in the lecture, we will try to switch to use better resources if the queuing situation becomes really bad close to the deadline.
It seems that using some of the more complex models with complex prompts, and generating multiple images causes the GPUs to run out of VRAM. I was able to mitigate this by only generating one image at a time, but even this sometimes returned errors, which throws an error for everything else in the queue as well.
I see...thanks for the feedback. I will discuss with the teaching team and see if we can have a better setup.
Good news: we have upgraded to a faster GPU with more VRAMs. Feel free to give it a try!
Its still very slow, taking about 10 minutes to generate one image. Can something be done about this? Because I may not be able to finish this assignment on time given the amount of images we have to generate.
See #12
I've been trying to do this assignment for a while, but the images generation takes a very long time, even though I'm the only one in the queue (I can't imagine how worse it'll get as the deadline approaches), and I keep getting errors.
Any advice would be appreciated!