JoyAI-Image-Edit-Plus: Multi-Image Instruction-Guided Editing by JD Open Source

JoyAI-Image-Edit-Plus is JD Open Source's multi-image instruction-guided editing model for 1-6 reference images, combining 8B MLLM + 16B MMDiT. BF16 and INT8 formats with native ComfyUI support."

J

JoyAI-Image-Edit-Plus

Image EditingMulti-ImageInstruction-GuidedCompositionMultimodal

Multi-image instruction-guided editing model by JD Open Source. Accepts 1 to 6 reference images along with a text instruction and generates a new image combining elements from all references. Natively supported in ComfyUI with pre-packaged model files.

DeveloperJD Open Source (JD.com)
Release Date2026-05
Architecture8B MLLM + 16B MMDiT
LicenseApache-2.0
ComfyUI SupportNative (since June 9, 2026)
FormatsBF16 (full precision) / INT8 ConvRot (quantized)
Reference Images1-6

Model Files

FileDirectoryFormat
joyai_image_edit_plus_bf16.safetensorsdiffusion_models/BF16
joyai_image_edit_plus_int8_convrot.safetensorsdiffusion_models/INT8 ConvRot
qwen3vl_8b_joyimage_edit_plus_bf16.safetensorstext_encoders/BF16
qwen3vl_8b_joyimage_edit_plus_int8_convrot.safetensorstext_encoders/INT8 ConvRot
wan_2.1_vae.safetensorsvae/BF16

Installation

Place the downloaded files in the following ComfyUI directory structure:

ComfyUI/
└── models/
    ├── diffusion_models/
    │   └── joyai_image_edit_plus_bf16.safetensors
    │   └── joyai_image_edit_plus_int8_convrot.safetensors
    ├── text_encoders/
    │   ├── qwen3vl_8b_joyimage_edit_plus_bf16.safetensors
    │   └── qwen3vl_8b_joyimage_edit_plus_int8_convrot.safetensors
    └── vae/
        └── wan_2.1_vae.safetensors

ComfyUI Workflow

A pre-built ComfyUI workflow is bundled in the HF repository:

FileLink
workflow_joyimage_edit_plus.jsonDownload

Guides and workflows related to this model series.

No articles found.

Comments

Sign in with GitHub to join the discussion.

Loading comments…