# LLaDA-Image: A Fully Open 6B Image Generator

> inclusionAI pairs a diffusion transformer with a frozen vision-language model and publishes the entire training recipe.

Published by The Open Weights on Sep 2, 2026. Canonical: https://theopenweights.com/news/llada-image-4mh6

## Key facts

- Company: inclusionAI
- Model: LLaDA-Image
- Version: LLaDA-Image
- Category: Text → Image
- Modalities: Image Editing, Text → Image
- License: Other (Open weights, commercial use allowed)
- Parameters: 6B
- Significance: notable
- Published: 2026-09-02
- Last verified: 2026-09-11
- Canonical URL: https://theopenweights.com/news/llada-image-4mh6

inclusionAI has released **LLaDA-Image**, a 6-billion-parameter diffusion transformer for text-to-image generation and image editing, along with a rare full accounting of how it was built. According to the [accompanying paper](https://huggingface.co/papers/2609.03796), the project's central claim is not just the model itself but a completely open training recipe — the kind of detail most labs keep behind closed doors.

The architecture leans on a frozen vision-language model rather than training a bespoke text encoder from scratch. That design keeps the language-understanding component fixed while the diffusion transformer learns to turn prompts into pixels, a strategy meant to make strong results reproducible without an enormous compute budget.

## Why it matters

Open weights are common; open recipes are not. By documenting the data pipeline, training procedure, and design choices end to end, LLaDA-Image gives researchers and smaller teams a template they can actually follow and modify.

- **6B diffusion transformer** covering both text-to-image and image editing
- **Frozen VLM** for prompt understanding, reducing training complexity
- **Distilled few-step variant** for faster inference

The distilled version is the practical hook here: few-step sampling cuts the generation latency that makes diffusion models expensive to serve. Combined with a mid-sized parameter count and a permissive-enough footprint, LLaDA-Image positions itself as a workable base for teams that want to study or extend open image generation rather than treat it as a black box.

## Get the model

- [HF Papers](https://huggingface.co/papers/2609.03796)
- [Hugging Face](https://huggingface.co/inclusionAI/LLaDA-Image)

## Sources

- [LLaDA-Image: Building Strong Image Generators with Fully Open Training Recipes](https://huggingface.co/papers/2609.03796) — HF Papers, Sep 2, 2026
- [inclusionAI/LLaDA-Image](https://huggingface.co/inclusionAI/LLaDA-Image) — Hugging Face, Aug 28, 2026

---
Source: The Open Weights (https://theopenweights.com/). Aggregated and written by Claude, curated by humans. Cite as: "LLaDA-Image: A Fully Open 6B Image Generator", The Open Weights, Sep 2, 2026, https://theopenweights.com/news/llada-image-4mh6