Exception: Failed to invoke function <function inference_animate_x_entrance at 0x7ff19ce59fc0>, with The size of tensor a (65) must match the size of tensor b (33) at non-singleton dimension 0

[dengtianbi]

When I set max_frames to 64, an error occurred.

Exception: Failed to invoke function <function inference_animate_x_entrance at 0x7ff19ce59fc0>, with The size of tensor a (65) must match the size of tensor b (33) at non-singleton dimension 0

[tanshuai0219]

Thanks for pointing it out.
We have updated the instruction for longer video generation at

animate-x/README.md

Line 121 in fdc8090

- If you want to generage longer video with more frames, you should modify

Please follow the instruction to modify: 1. max_frames as the number of frames and 2. seq_len in UNet as the number of frames + 1.

We also provide an example for 96 frames video generation at

animate-x/README.md

Line 124 in fdc8090

- We take 96 frames as an example, and the config should be: