This is a multi-image fusion generation application based on the FLUX.2 Klein dual-reference control chain. Users upload two reference images (subject image + auxiliary image) and enter a synthesis prompt. The system simultaneously utilizes the latent information of both reference images for generation, achieving stable fusion of characters, structures, and styles. Compared to standard single-reference generation, this pipeline introduces dual ReferenceLatent control + dual Ref Controller + Text/Ref Balance + Color Anchor, allowing "structural control, prompt control, and color control" to be adjusted independently, significantly enhancing the stability and controllability of dual-image synthesis.
It is especially suitable for scenarios such as: dual-person portrait generation, character synthesis, face + pose combination, character + scene fusion, and style transfer overlay. The core advantages are: both reference images stably participate in the generation without only one taking effect, while colors do not drift, structures remain intact, and prompts can still modify the image. If you want it to look more like the reference images, you can increase the structural strength; if you want it to follow the prompts more closely, you can increase the text weight; if color distortion occurs, you can decrease the color anchor strength.
🎁Claim RH Coins first, then experience the workflow! Click the avatar in the top right corner → Invitation Code → Enter [rh-v1111] to instantly receive 1000 RH Coins, and log in daily to get more RH Coins~ For more ComfyUI workflows, tutorials, and gameplay, follow our official account (AIKSK), with simultaneous updates on Douyin / Bilibili / Xiaohongshu / YouTube (AI-KSK) ✨


No creations yet

No creations available.