baichuan2 7b-13b cannot be 8-bit weight quantized. for baichuan2 7b, the error is: torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 40.00 GiB. GPU 0 ...
Due to python-pillow/Pillow#7568, some local GIF headers are no longer omitted with pillow 10.2.0, and this breaks our very simple kraken3 screen smoke test ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
