Some images from MiniMax H3 Image
retained text
When I've seen the results, I've found them kind of outdated, but I know that a lot of people in the AI image gen community like the old look of the early days of AI image gen. More images with prompts are on my [Substack](https://h2logs.substack.com/p/minimax-h3-image-showcase-and-workflow). I've built an image-to-image workflow on [Civitai](https://civitai.com/models/3049181/minimax-h3-image-workflow-image-to-image).
document fetched 2026-11-14 00:31 · canonical https://reddit.com/r/StableDiffusion/comments/1xvcede/some_images_from_minimax_h3_image