I'm working on a few fashion ideas and was looking for one of these tools for a while. The ability to mockup samples is pretty amazing. You can upload flat lays and pretty easily make try-on models with various templates and backgrounds. Swapping fashion models also works well. The images are downloadable in UHD*.

NOTE: When you generate an image and view it, there is an icon for "image to video" (took me a while to find it). Before I saw that, I was downloading and then re-upload pictures to make showcase videos.

FUTURE NICE TO HAVES:
1. I hope they include bulk processing (heard its in the roadmap).
2. There should be a selection option on the main video screen to generate video from existing image (as noted above, I only found this option by opening an image first, but couldn't find it from the video screen).
3. It would be nice to have a linked media library for all of the generation models. Right now, it appears all of the different generation tools (swap fashion model, virtual try-on, video creation etc) can only use images uploaded into that specific tool. Its kind of annoying having to upload the same image 5x times in a row.

PROBLEMS: The only problem I've encountered so far is when using the virtual accessories try-on mode and generating mockups of products that have text on it. I found the AI often transposes the text to a different part of the product or makes the font unreadable. This is expected though since pretty much every current AI model as similar problems. I probably had 1/15 good images depending on the model and product with text on it.

*UHD supports output size of 4096*4096. If the image size already exceeds 4096*4096, original-size image will be exported.