When I try to generate videos by using this model, I found that some case lead to a blurry output video (Not because of 480p, but the main part of video is blurry). Are there something wrong with my parameters or other reason?
A young person with multicolored hair, wearing a dark outfit adorned with gold and blue tassels. Outdoor scene with a sunlit forest in the background. Her eyes widened slightly, and her mouth parted just a little, capturing an expression of mild surprise. There was a flicker of something else in her gaze—maybe curiosity, maybe a hint of delight—like she had stumbled upon something unexpectedly wonderful.
A young person in a black sweater with a necklace and dark hair. Indoor scene with a softly lit background. He raised his hand and gave a small wave, his lips curling into a faint smile. His eyes were warm, and there was a subtle nod of acknowledgment, the gesture calm and unhurried, radiating a quiet friendliness.
A young man in a black coat with intricate designs and long red hair. Indoor scene with a soft, warm light in the background. His eyes widened slightly, and his mouth parted just a little, a subtle expression of surprise washing over his face. There was a hint of excitement in his gaze, as if something had unexpectedly delighted him, leaving him momentarily caught between astonishment and joy.
Hi, Thanks for your release.
When I try to generate videos by using this model, I found that some case lead to a blurry output video (Not because of 480p, but the main part of video is blurry). Are there something wrong with my parameters or other reason?
runing scripts(no modify): src/mochi_preview/gradio_ui.py devices: A100(40GB)*8 CUDA version: 11.8 python version: 3.10.15 torch version: 2.4.1+cu118
generate parameters:
Prompts:
https://github.com/user-attachments/assets/f70b627e-b282-4b4a-a77b-e838321e6e69
https://github.com/user-attachments/assets/f5fc5cfc-9f62-48f2-929b-00098a7f2037
https://github.com/user-attachments/assets/c275aa6c-b084-4ef1-9486-81d4a4e4531f