<p>🇬🇧English:<br />This MOSS-Image ComfyUI workflow brings together automatic prompt expansion, reference-image captioning, image generation, and SeedVR2 upscaling. Start with a short description and let a local Qwen vision-language model expand it into a detailed prompt, or upload a reference image and combine its visual description with your requested changes to create a new interpretation. Customize prompt prefixes, LoRAs, aspect ratios, and resolution, then compare the original output with the upscaled result. Suitable for text-to-image creation, reference-based reinterpretation, and style exploration.</p><p>🇨🇳中文：<br />这是一套基于 MOSS-Image 的 ComfyUI 自动化工作流，将提示词扩写、参考图反推、图像生成与 SeedVR2 放大整合在一起。输入一句简短描述，即可通过本地 Qwen 视觉语言模型扩展画面细节并生成图片；也可以上传参考图，提取画面描述，结合修改要求进行重新创作。支持自定义提示词前缀、LoRA、画幅和分辨率，并提供放大前后对比，适合文生图、参考图再创作和风格探索。</p><div data-youtube-video><iframe width="640" height="480" allowfullscreen="true" autoplay="false" disablekbcontrols="false" enableiframeapi="false" endtime="0" ivloadpolicy="0" loop="false" modestbranding="false" origin playlist src="https://www.youtube.com/embed/8UffpngDEVk?rel=1" start="0"></iframe></div><p></p>

Files with SHA256: 6c08ddb61d788279a55a036688debf2ed4fe2a35942e69ef64d925a1d356d44e