search

PKU-YuanGroup / Open-Sora-Plan

This project aim to reproduce Sora (Open AI T2V model), we wish the open source community contribute to this project.
Apache License 2.0
10.87k stars 971 forks source link

When I use Open-Sora-Plan-v1.1.0 inference, I get a bunch of noise #320

Open yiiizuo opened 1 week ago

yiiizuo commented 1 week ago

When I use Open-Sora-Plan-v1.1.0 inference, I get a bunch of noise。

CUDA_VISIBLE_DEVICES=1 python opensora/sample/sample_t2v.py \ --model_path /zuoyi/T2V/Open-Sora-Plan-v1.1.0 \ --version 65x512x512 \ --num_frames 65 \ --height 512 \ --width 512 \ --cache_dir "./cache_dir" \ --text_encoder_name /zuoyi/T2V/t5-v1_1-xxl \ --text_prompt examples/prompt_list_0.txt \ --ae CausalVAEModel_4x8x8 \ --ae_path "/zuoyi/T2V/Open-Sora-Plan-v1.1.0" \ --save_img_path "./sample_video_65x512x512" \ --fps 24 \ --guidance_scale 7.5 \ --num_sampling_steps 150 \ --enable_tiling

https://github.com/PKU-YuanGroup/Open-Sora-Plan/assets/44500387/73d5ef68-ea15-4e96-a775-155633609b03

0