jdopensource/JoyAI-Image-Edit-ComfyUI

JoyAI-Image-Edit (ComfyUI weights)

Single-file .safetensors checkpoints of JoyAI-Image-Edit, repackaged for the standard ComfyUI model folder layout.

Use these with the joyai_image_comfyui custom node from the JoyAI-Image GitHub repo.

Files

File	Size	Goes into	Component
diffusion_models/joy_image_edit_bf16.safetensors	~32 GB	ComfyUI/models/diffusion_models/	JoyImageEditTransformer3DModel (bf16)
text_encoders/qwen3vl_joyimage_bf16.safetensors	~17 GB	ComfyUI/models/text_encoders/	Qwen3VL text encoder (bf16)
vae/joy_image_edit_vae.safetensors	~254 MB	ComfyUI/models/vae/	AutoencoderKLWan

The repo's directory layout already matches ComfyUI/models/, so a single hf download into your models root drops every file where it needs to go.

Quick start

# 1) Install the custom node
cd ComfyUI/custom_nodes
git clone https://github.com/jd-opensource/JoyAI-Image.git
cp -r JoyAI-Image/joyai_image_comfyui ./
rm -rf JoyAI-Image

# 2) Download the weights straight into ComfyUI/models/
hf download jdopensource/JoyAI-Image-Edit-ComfyUI \
  --local-dir /path/to/ComfyUI/models

# 3) Restart ComfyUI — new nodes appear under `loaders/joyai` and `image/joyai`.

Requirements

• diffusers >= 0.39.0.dev0 (install from source: pip install git+https://github.com/huggingface/diffusers.git)
• transformers >= 4.57.0

GGUF quantizations

Lower-bit GGUF quants of the transformer and text encoder are available at huangfeice/JoyAI-Image-Edit-Diffusers-GGUF (community contribution). The VAE here is the only VAE you need — GGUF doesn't quantize VAE.

Links

• Source code, node implementation, and full documentation: github.com/jd-opensource/JoyAI-Image
• Original Diffusers-format weights: jdopensource/JoyAI-Image-Edit-Diffusers