ali-vilab

/

text-to-video-ms-1.7b

like 582

Follow

ali-vilab 38

well, its solved, I had to restart it but the crash helped. now another please wish. my GPU is only pure 10GB and gets out of memory for a long one, is there any CPU solution?

I am on win11 can you please tell me where is saving the video?

me too

Add this code at the end:
print(f"PATH: {video_path}")
or
print(video_path)

Add this code at the end:
print(f"PATH: {video_path}")
or
print(video_path)

The video_path value can be printed out, but the file does not exist

did you find solution? I'm experiencing the same on win11

print(f"PATH: {video_path}

Worked. It was saving vids to temp folder.

Would love to know how to pick destination folder. I'm sure its not too hard.

Would love to know how to pick destination folder. I'm sure its not too hard.

I use something like this to save the video in my "Videos" folder with a specified file name:

import os
import torch
from diffusers import DiffusionPipeline, DPMSolverMultistepScheduler
from diffusers.utils import export_to_video

pipe = DiffusionPipeline.from_pretrained("damo-vilab/text-to-video-ms-1.7b", torch_dtype=torch.float16, variant="fp16")
pipe.scheduler = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config)
pipe.enable_model_cpu_offload()

prompt = "Spiderman is surfing"
video_frames = pipe(prompt, num_inference_steps=25).frames
video_name = "spiderman_surfing.mp4"
video_path = os.path.join(os.path.expanduser('~'), 'Videos', video_name)

export_to_video(video_frames, video_path)