and r1 beats gpt o4 often in academic scenario
and r1 beats gpt o4 often in academic scenario
9 Replies
can we speed up sdxl lora training using tensor RT?
they are amazing
open ai and meta already worried
nope
tensorrt only for inference
Anyone know why this error is all I get recently? There may have been an update, but I'm not sure why the HASH would be different than expected..
Even copying a working install of forge from another pc doesn't fix this. Something is broken as far as torch goes it seems..
Ok, figured out a fix. Delete pip's cache directory and make it download fresh installs. Seems to have gotten around that error.
Local alternatives to ideogram 2.0? Looking for good text, canvas ability where you can edit select parts of a generation with a prompt, and ability to upload images to remix too. Anything less than $20 a month haha
isp / computer related
also delete your pip cache
pip cache purge
python -m pip install --upgrade pip
haha i see you found 😄
i dont think such exists :d
EMO2
EMO2: End-Effector Guided Audio-Driven Avatar Video Generation.
Wouldnt get too excited no code released and its our old friends Alibaba.....I dont think they even released EMO1
looks fake
but no code :d