This is an Image-to-Image enhancement workflow based on ERNIE-Image Turbo.
Users upload an image, and combined with prompts, the model redraws and optimizes details based on the original image. The entire workflow structure is very standard:
The image first goes through resolution control, is encoded into latent, then combined with prompts through the sampler to generate a new image, and finally decoded via VAE to output the final image.
Compared to standard Text-to-Image, the features of this workflow are:
Applicable scenarios:
🎁 Claim RH Coins first, then try out the workflow! Click your avatar in the top right corner → Invitation Code → Enter [rh-v1111] to instantly receive 1000 RH Coins. Log in daily to claim another 100 coins~ For more ComfyUI workflows, tutorials, and tips, follow our WeChat Official Account (AIKSK). Synchronized updates also available on TikTok / Bilibili / Xiaohongshu / YouTube (AI-KSK) ✨


No creations yet

No creations available.