Bill Zhao

求一个quantize好的模型

3

不是显存不够，是内存也不够，无法先加载模型再进行quantize. 有没有quantize 8bit或者4bit的教程？谢谢

可不可以试试visualglm?

1

感觉这个对visualglm 这种吃内存显存的更有用

Does the integration with diffusers support stable diffusion XL model?

Img2txt result is pretty bad on 16bit

4

Don't know if I did anything wrong, but the result was not like the example. import os from core.models.model_module_infer import model_module model_load_paths = ['CoDi_encoders.pth', 'CoDi_text_diffuser.pth', 'CoDi_audio_diffuser_m.pth', 'CoDi_video_diffuser_8frames.pth'] inference_tester = model_module(data_dir='checkpoints/',...

ValueError: unet/hotshot_xl.py as defined in `model_index.json` does not exist in hotshotco/Hotshot-XL

5

diffusers/pipelines/pipeline_utils.py", line 1680, in download """ ValueError: unet/hotshot_xl.py as defined in `model_index.json` does not exist in hotshotco/Hotshot-XL and is not a module in 'diffusers/pipelines'. Somehow, when I first launched the...

The error happens when loading checkpoint shards: 50%|████████ | 4/8 python3.10/site-packages/torch/serialization.py: ││ 283 in __init__ ││ ││ 280 ││ 281 class _open_zipfile_reader(_opener): ││ 282 │ def __init__(self, name_or_buffer) -> None:...

I have tried different ways but unfortunately don’t know how to keep alpha channel for muiltiview hevc. Do you have any ideas?