蚂蚁 inclusionAI 开源 LLaDA-Image 图像生成与编辑模型

inclusionAI; LLaDA-Image

Repository navigation

NotificationsYou must be signed in to change notification settings; Fork12 (12); Star261 (261)You must be signed in to star a repository

Repository files navigation

README

Welcome to the official repository for LLaDA-Image, a unified model for high-quality image generation and editing.

Photorealistic image generation with natural lighting, lifelike details, and coherent scenes.

High-quality text rendering and creative poster generation across diverse visual styles.

Instruction-guided image editing with faithful content preservation and precise visual changes.

Introduction

LLaDA-Image is a competitive 6B-parameter open-source unified image generation and editing model family. It includes LLaDA-Image, a 50-step Base model for high-quality text-to-image generation and instruction-guided editing, and LLaDA-Image-Turbo, a 4-step distilled model for fast generation and editing. Both variants support practical text-to-image generation, VQ-conditioned generation, reference-image editing, and Chinese--English text rendering.

This repository provides the checkpoints and Diffusers-based inference code for the LLaDA-Image model family.

2026-09-07: Thanks to @Rebel AI, the Comfy UI for LLaDA-Image and LLaDA-Image-Turbo and its FB8 version is supported; 2026-09-04: We released the LLaDA-Image Base and Turbo checkpoints together with the inference code

Highlights

Unified generation and editing. A single checkpoint supports text-to-image generation and reference-preserving, instruction-guided editing without a separate editing backbone; Unified diffusion model. Both backbone and DiT are diffusion models, trained in a unified framework; Realistic image generation. LLaDA-Image produces high-quality images with rich visual details, natural lighting, and coherent compositions; Image-only pre-training for visual-prior learning. The report establishes the visual prior through image-only pre-training and mid-training before introducing paired language supervision and joint generation--editing training; Efficient inference with distilled model. LLaDA-Image-Turbo uses Twin-DMD distillation to deliver fast image generation and editing in only 2--4 sampling steps; SOTA on Qwen-Image-Bench. LLaDA-Image achieves state-of-the-art overall scores of 53.53 in English and 53.38 in Chinese

Model Zoo

Opensource Plan

Inference code and model weights; Training code (coming soon)

Quick Start

1. Create an environment

The implementation has been used with Python 3.11, PyTorch 2.8, Transformers 4.57.6, and Diffusers 0.39.0.

2. Run inference

The pipeline accepts a prompt and, for editing, an optional reference image.

LLaDA-Image (Base)

Use the Base checkpoint for high-fidelity generation and editing. Its recommended sampling configuration is 50 steps.

LLaDA-Image-Turbo

Use the Turbo checkpoint for fast generation and editing. Its recommended sampling configuration is 4 steps.

Note

For LLaDA-Image-Turbo inference, you can try setting stochastic_sampling to false in scheduler/scheduler_config.json, which may produce sharper details in some cases.

Generation modes

Both checkpoints support the following modes. Text and VQ-conditioned generation require height and width divisible by 16; image editing requires dimensions divisible by 32.

VQ-conditioned generation uses the LLaDA2 model to produce image VQ tokens from the prompt, which SigVQ embeds before diffusion. Do not provide an input image in VQ mode.

Image editing requires a reference image:

Acknowledgements

We thank the VeOmni project and its contributors for their valuable open-source work.

Citation

If you find LLaDA-Image useful for your research or applications, please consider citing our work.