-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathDockerfile
More file actions
18 lines (15 loc) · 1.51 KB
/
Dockerfile
File metadata and controls
18 lines (15 loc) · 1.51 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
# clean base image containing only comfyui, comfy-cli and comfyui-manager
FROM runpod/worker-comfyui:5.5.1-base
# install custom nodes into comfyui (first node with --mode remote to fetch updated cache)
# No registry-verified custom nodes found.
# Unknown registry nodes present but none provided an aux_id (GitHub repo) so they could not be resolved:
# - MarkdownNote
# - MarkdownNote
# - MarkdownNote
# download models into comfyui
RUN comfy model download --url https://huggingface.co/Comfy-Org/Qwen-Image-Edit_ComfyUI/resolve/main/split_files/diffusion_models/qwen_image_edit_2511_bf16.safetensors --relative-path models/diffusion_models --filename qwen_image_edit_2511_bf16.safetensors
RUN comfy model download --url https://huggingface.co/lightx2v/Qwen-Image-Edit-2511-Lightning/resolve/main/Qwen-Image-Edit-2511-Lightning-4steps-V1.0-bf16.safetensors --relative-path models/loras --filename Qwen-Image-Edit-2511-Lightning-4steps-V1.0-bf16.safetensors
RUN comfy model download --url https://huggingface.co/Comfy-Org/HunyuanVideo_1.5_repackaged/resolve/main/split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors --relative-path models/text_encoders --filename qwen_2.5_vl_7b_fp8_scaled.safetensors
RUN comfy model download --url https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/resolve/main/split_files/vae/qwen_image_vae.safetensors --relative-path models/vae --filename qwen_image_vae.safetensors
# copy all input data (like images or videos) into comfyui (uncomment and adjust if needed)
# COPY input/ /comfyui/input/