deepbeepmeep

I have been succesful with a RTX 4090 by enabling VAE slicing and CPU offloading. Add at line 545 of gradio_app_sdxl_specific_id.py: [TAB]pipe.enable_vae_slicing() [TAB]pipe.enable_model_cpu_offload() You may need to use an old...

It would be very useful to edit a mask and still be able to see behind it. This should be simple to do as the RGBA format is already supported...

thanks. How do you set "attn_impl: torch" ?

ok I have figured how to do it. Here are the results: - with Windows, it hangs forever after 'the "Warming up.." message. The reason could be that it is...

Fyi Triton 2.1 supports ADA while the current version you use (2.0) doesn't. llm foundry won't le met use Triton 2.1 since version 2.0 is hardcoded.

Triton 2.0 is hard coded inside the Hugging Face transfomer code. So there is no way I can change it. Besides my main problem right now is that the initialization...

Hi, I would be grateful if you could provide a fix also for the local LLM since it is broken as well