JoyAI-Image-Edit: Single-Image Instruction-Guided Editing by JD Open Source
JoyAI-Image-Edit is JD Open Source's single-image instruction-guided editing model, combining 8B MLLM + 16B MMDiT. Available in BF16 and INT8 formats with native ComfyUI support.
J
JoyAI-Image-Edit
Image EditingInstruction-GuidedSpatial EditingMultimodalSingle-image instruction-guided editing model by JD Open Source. Accepts an input image and natural language instruction and generates the edited result. Natively supported in ComfyUI with pre-packaged model files in BF16 and INT8 ConvRot formats.
| Developer | JD Open Source (JD.com) |
| Release Date | 2026-05 |
| Architecture | 8B MLLM + 16B MMDiT |
| License | Apache-2.0 |
| ComfyUI Support | Native (since June 9, 2026) |
| Formats | BF16 (full precision) / INT8 ConvRot (quantized) |
Model Files
| File | Directory | Format |
|---|---|---|
joyai_image_edit_bf16.safetensors | diffusion_models/ | BF16 |
joyai_image_edit_int8_convrot.safetensors | diffusion_models/ | INT8 ConvRot |
qwen3vl_8b_joyimage_edit_bf16.safetensors | text_encoders/ | BF16 |
qwen3vl_8b_joyimage_edit_int8_convrot.safetensors | text_encoders/ | INT8 ConvRot |
wan_2.1_vae.safetensors | vae/ | BF16 |
Installation
Place the downloaded files in the following ComfyUI directory structure:
ComfyUI/
└── models/
├── diffusion_models/
│ ├── joyai_image_edit_bf16.safetensors
│ └── joyai_image_edit_int8_convrot.safetensors
├── text_encoders/
│ ├── qwen3vl_8b_joyimage_edit_bf16.safetensors
│ └── qwen3vl_8b_joyimage_edit_int8_convrot.safetensors
└── vae/
└── wan_2.1_vae.safetensorsComfyUI Workflow
A pre-built ComfyUI workflow is bundled in the HF repository:
| File | Link |
|---|---|
workflow_joyimage_edit.json | Download |
Related News
Guides and workflows related to this model series.
No articles found.
Comments
Sign in with GitHub to join the discussion.