使用CogVideoX 1.5运行Gradio Demo时出现张量不匹配

[yschdxm]

System Info / 系統信息

Cuda：12.9 Diffusers：0.34.0 Python：3.10.18 系统：Windows 11 专业版 24H2 26100.4484

Information / 问题信息

• [x] The official example scripts / 官方的示例脚本
• [ ] My own modified scripts / 我自己修改的脚本和任务

Reproduction / 复现过程

运行Gradio Demo并使用I2V生成视频（使用任何图片），发生以下报错（张量不匹配）

发生异常: RuntimeError Sizes of tensors must match except in dimension 1. Expected size 16 but got size 8 for tensor number 1 in the list. File "F:\UPC\YNU\CogVideo\inference\gradio_composite_demo\app.py", line 252, in infer video_pt = pipe_image( File "F:\UPC\YNU\CogVideo\inference\gradio_composite_demo\app.py", line 464, in generate latents, seed = infer( RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 16 but got size 8 for tensor number 1 in the list.

Expected behavior / 期待表现

正常生成视频