This was an exploration of the capabilities of the MiniMax H3 video model, and the results were surprisingly good. It fully retained the advantages of the H3 video model, such as consistency, while also solving many image editing problems. Editing large-angle images becomes very simple and has a very high success rate, almost always getting it right on the first try. Therefore, I highly recommend everyone to try it. For details, you can refer to the YouTube video, where I explain things in great detail. You can also directly use my workflow to experience it.
🔗 workflow online
https://www.runninghub.ai/zh-cn/post/2120476647413985281/?inviteCode=rh-v1256
👍 The API platform i'm using
e.g. Minimax h3 15s 1440p ¥0.3($0.05)

