Inside Flux 3 Image Klein with BFL!

We're going behind the scenes of how FLUX 3 Image was built with special guest Yumeng Li (Research Scientist at Black Forest Labs) and Paul Hansen (Creative Technologist at ComfyUI) for a live conversation covering the research decisions, the tradeoffs, and what it took to build a model that lets you place every element exactly where you want it. FLUX 3 Image lets you compose a scene with bounding boxes, describe what goes in each one, and edit a finished image one box at a time while everything you didn’t touch stays put. What we’ll cover: 🎨 The story behind FLUX 3 Image, from research to release 🎨 Composing with bounding boxes: layout prompts, element tables, and how the model reads them 🎨 Pixel-perfect editing, with multiple targeted edits at once 🎨 Multi-reference generation with up to 10 images, and native 2K and 4K output 🎨 Why it’s built for agents, and what that means for your workflows 🤑 How to use it in a special ComfyUI Competition with $25k worth in prizes 🎨 Live Q&A Drop your questions in chat and we'll answer them live! ⏰ Subscribe and turn on notifications to catch the stream live 💬 Submit questions in chat during the broadcast or in the comments before we go live 👍 Leave a like if you want more behind-the-model conversations Comfy.org https://comfy.org/ https://comfy.org/workflows