(For SOTAI Squad) USO Style Transfer Detail Tutorial in ComfyUI | Crazy Results on Just 8GB VRAM

Link video: https://www.youtube.com/watch?v=PD_yc1Pbmjc

In this video, I’ll show you how to set up and run the USO workflow to create stunning, consistent style transfer — and the best part? It runs smoothly on just 8GB VRAM ?.

• What you’ll learn:

  • How to install & set up the USO workflow in ComfyUI

  • Tips to get consistent results with style transfer

  • How to optimize for low VRAM GPUs (8GB)

• Link models:

  • flux1-dev-Q3_K_S.gguf : https://huggingface.co/city96/FLUX.1-dev-gguf/resolve/main/flux1-dev-Q3_K_S.gguf?download=true

    • Put to models/unet folder

  • t5xxl-Q3_K_S.gguf: https://huggingface.co/city96/t5-v1_1-xxl-encoder-gguf/resolve/main/t5-v1_1-xxl-encoder-Q3_K_S.gguf?download=true

    • Put to models/clip folder

  • clip_l.safetensor : https://huggingface.co/comfyanonymous/flux_text_encoders/resolve/main/clip_l.safetensors

    • Put to models/clip folder

  • ae.safetensor: https://huggingface.co/black-forest-labs/FLUX.1-dev/resolve/main/ae.safetensors

    • Put to models/vae folder

  • uso-flux1-dit-lora-v1 : https://huggingface.co/Comfy-Org/USO_1.0_Repackaged/resolve/main/split_files/loras/uso-flux1-dit-lora-v1.safetensors

    • Put to models/loras folder

  • uso-flux1-projector-v1.safetensors : https://huggingface.co/Comfy-Org/USO_1.0_Repackaged/resolve/main/split_files/model_patches/uso-flux1-projector-v1.safetensors

    • Put to models/model_patches folder

  • sigclip_vision_patch14_384.safetensors: https://huggingface.co/Comfy-Org/sigclip_vision_384/resolve/main/sigclip_vision_patch14_384.safetensors

    • Put to models/clip_vision folder

  Note:

  • Install custom node "ComfyUI_LayerStyle" to get "Purge VRAM" node

• Please check the workflow in the attachments.