开源图像模型首次全面公开训练配方
开源图像模型第一次把「怎么练出来的」全盘托出：6B 参数的扩散模型，先用 2.2 亿张纯图片自己学视觉规律，再配一个冻结的视觉语言模块，最后用 Muon 优化器加速。结果在英文和中文的 Qwen-Image-Bench 上都拿了开源第一，还蒸馏出一个 2-4 步出图的快速版。它不是你明天能直接用的工具，但它是开源社区的分水岭——以前你只能拿到权重，现在连训练细节都能复现，意味着小团队也能站在同一块地基上往上盖。
📄 原文摘要(英文)
We introduce LLaDA-Image, a unified framework that pairs a 6B Diffusion Transformer (DiT) trained from scratch with a frozen vision-language understanding module built on the LLaDA2.0-Mini diffusion language model backbone. Instead of relying heavily on paired image-text data from the beginning, we first build a strong visual generative prior through image-only pre-training and mid-training. The generation pipeline comprises 220M samples, 98 of which are real images. For efficient and scalable optimization, we use parameter-free RMSNorm throughout the DiT together with the Muon optimizer. The resulting unified model produces highly photorealistic images while accurately following fine-grained editing instructions. We further distill LLaDA-Image into LLaDA-Image-Turbo, enabling fast inference in 2-4 sampling steps. On Qwen-Image-Bench, LLaDA-Image achieves overall scores of 53.53 and 53.38 on the English and Chinese tracks, respectively, setting a new state-of-the-art among open-source models on both tracks. To support further research on capable and efficient generative models, we release our model weights, training code, and detailed recipes.