Update README.md

Update README.md

norm

Update preprocess.bash

Add sweep integration

Implement bulk prompting, fast model in CLI

Add fast-dit download script

Update README.md to include a diffusers trainer note (#118)

Fix assertion on multi-GPU

add longer timeout to preprocess (#105)

Fix missing qkv_x, qkv_y layers. Remove unsupported qkv_proj_lora_mask argument to simplify (#103)

Add --cpu_offload to README for inference scripts (#102)

Fix Gradio UI, adding supoort for LoRA inference. Change default CFG scale (#101)

Support non-LoRA model construction (#100)

Fix single GPU encoding (#99)