lucataco/video-crafter 🔢📝 → 🖼️

▶️ 10.8K runs 📅 Oct 2023 ⚙️ Cog 0.8.6 🔗 GitHub 📄 Paper

text-to-video

About

Open diffusion model for high-quality video generation

Example Output

Prompt:

"an astronaut running away from a dust storm on the surface of the moon, the astronaut is running towards the camera, cinematic"

Output

Performance Metrics

189.22s Prediction Time

331.45s Total Time

All Input Parameters

{
  "fps": 15,
  "seed": 64045,
  "steps": 40,
  "width": 1024,
  "height": 576,
  "prompt": "an astronaut running away from a dust storm on the surface of the moon, the astronaut is running towards the camera, cinematic"
}

Input Parameters

fps Type: integerDefault: 28: Frames per second of the output video
seed Type: integer: Random seed. Leave blank to randomize the seed
steps Type: integerDefault: 50: Number of steps to take in the video
width Type: integerDefault: 1024: Width of the output video
height Type: integerDefault: 576: Height of the output video
prompt (required) Type: string: Input Prompt

Output Schema

Output

Type: string • Format: uri

Example Execution Logs

@CoLVDM Inference: 2023-10-19-04-17-09
Global seed set to 64045
AE working on z of shape (1, 4, 64, 64) = 16384 dimensions.
>>> model checkpoint loaded.
[rank:0] 1/1 samples loaded.
[rank:0] batch-1 (1)x1 ...
DDIM scale True
ddim device cuda:0
/root/.pyenv/versions/3.9.18/lib/python3.9/site-packages/torch/utils/checkpoint.py:31: UserWarning: None of the inputs have requires_grad=True. Gradients will be None
warnings.warn("None of the inputs have requires_grad=True. Gradients will be None")
Saved in /VideoCrafter/results. Time used: 163.36 seconds

Version Details

Version ID: 7757c5775e962c618053e7df4343052a21075676d6234e8ede5fa67c9e43bce0
Version Created: October 19, 2023

Run on Replicate →