I'm Tutu. I started making AI images in 2023 after seeing a friend show me what AI image generation could already do. I was shocked. In my mind, AI images were still rough and low-quality, but what I saw was already close to professional illustration and photography. That pulled me into Midjourney, SD 1.5, Civitai, and later LoRA training. I spent a long time generating images, sharing them online, and training my own models. The deeper I went, the more painful the workflow became: hands were often broken, lighting and details were inconsistent, prompts were hard to organize, datasets needed endless cleanup, and captioning without subject filtering caused repeated pollution during training. So I started building my own tool. At first, my coding skills were rusty. I had to relearn a lot, study the basics again, and later AI coding helped me move much faster. Along the way, I also improved my own difference-extraction method and used it to create many free LoRA models. Those models have been downloaded and used millions of times globally. In 2025, after losing my job, I pushed even harder on building creator tools. The first version of Tutu AI Prompt Reverse Engineer launched on a small Chinese creator platform on April 30, 2025. Many friends and users supported it, gave feedback, and helped shape the product. Today, after 22 major updates, it has grown from a simple prompt-phrase captioning tool into a full Windows workspace for image and video prompt reverse engineering, batch annotation, prompt editing, dataset preparation, preprocessing, and image generation. I'm still building it every day, and my goal is simple: make it one of the best tools in this field for AI creators. I'd love to hear your feedback, especially if you work with image prompts, video prompts, LoRA datasets, or AI visual workflows.

Tutu AI Prompt Reverse Engineer

Reverse, edit, and manage image/video prompts

Reverse, edit, and manage image/video prompts