Follow these steps to set up the environment for your ComfyUI workflow:
1οΈβ£ Create a new virtual environment:
conda create -n [νλ‘μ νΈλͺ
] python=3.10
conda activate [νλ‘μ νΈλͺ
]
2οΈβ£ κ°μνκ²½ μμμ PyTorch with CUDA support μ€μΉ:
conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia
3οΈβ£ κ°μνκ²½ μμ± ν κ°μνκ²½ μμμ νμν νλ‘κ·Έλ¨ μ€μΉ:
requirements.txt
file and run:pip install -r requirements.txt
4οΈβ£ μ¬μ©νκ³ μΆμ λͺ¨λΈ λ€μ΄λ‘λ ν κ° λλ ν 리μ λ§κ² λ£μ΄μ€λ€.
ComfyUI/models/controlnet/
directory: control_v11p_sd15_lineart_fp16.safetensors5οΈβ£ νμν 컀μ€ν λ Έλ λ€μ΄λ‘λ :
custom_nodes
directory:cd ComfyUI/custom_nodes
git clone [ν΄λΉ 컀μ€ν
λ
Έλ git url]
6οΈβ£ checkpoint model λ€μ΄λ‘λ:
ComfyUI/models/checkpoints/
directory.7οΈβ£ Run ComfyUI:
python main.py
μ΄ νλ‘μ νΈλ₯Ό μ€ννκΈ° μν΄μλ λ€μ 컀μ€ν λ Έλλ€μ΄ νμν©λλ€.
μ΄ νλ‘μ νΈλ₯Ό μ€ννκΈ° μν΄μλ λ€μ λͺ¨λΈλ€μ΄ νμν©λλ€.