FramePack F1 video generated by Minshen, built using RH official flow, default generates 30-frame videos
Advantages: ① Stability is better compared to WAN (let alone Hunyuan...)
② Can generate infinitely, as long as the machine hardware meets the requirements, it can generate stably without crashing like WAN and Hunyuan after exceeding 6 seconds (inherent advantage of FramePack)
③ Generation efficiency is faster than WAN and does not depend on the resolution of reference images (WAN and Hunyuan both perform better with higher resolution input images)
Disadvantages: ① A bit too stable, dynamic effects are not very good, somewhat stiff (background dynamics, physical trajectories, etc. are inferior to WAN), better suited for scenes with less overall dynamics like indoor rooms
② The frame rate always feels half as slow, i.e., the generated motion feels very slow, 30 frames generated seem like 30 frames displayed as 15 frames (though it is actually 30 frames output). However, I have already applied frame rate doubling processing to avoid this
③ Understanding of prompts is not very good, the generated actions are always conservative relative to the prompts.
Creating is not easy. If this has helped you, please bookmark, like, and execute applications or workflows online more often. Everyone's support is my motivation for creating. Thank you for your support!
If you have questions, please leave a comment. I will reply promptly when I see it. Welcome to exchange ideas, learn together, and make progress together!
FramePack F1 video generated by Minshen, built using RH official flow, default generates 30-frame videos
Advantages: ① Stability is better compared to WAN (let alone Hunyuan...)
② Can generate infinitely, as long as the machine hardware meets the requirements, it can generate stably without crashing like WAN and Hunyuan after exceeding 6 seconds (inherent advantage of FramePack)
③ Generation efficiency is faster than WAN and does not depend on the resolution of reference images (WAN and Hunyuan both perform better with higher resolution input images)
Disadvantages: ① A bit too stable, dynamic effects are not very good, somewhat stiff (background dynamics, physical trajectories, etc. are inferior to WAN), better suited for scenes with less overall dynamics like indoor rooms
② The frame rate always feels half as slow, i.e., the generated motion feels very slow, 30 frames generated seem like 30 frames displayed as 15 frames (though it is actually 30 frames output). However, I have already applied frame rate doubling processing to avoid this
③ Understanding of prompts is not very good, the generated actions are always conservative relative to the prompts.
Creating is not easy. If this has helped you, please bookmark, like, and execute applications or workflows online more often. Everyone's support is my motivation for creating. Thank you for your support!
If you have questions, please leave a comment. I will reply promptly when I see it. Welcome to exchange ideas, learn together, and make progress together!


No creations yet

No creations available.