arthur-qiu/longercrafter 🔢📝❓ → 🖼️

▶️ 14.9K runs 📅 Oct 2023 ⚙️ Cog 0.8.3 🔗 GitHub 📄 Paper ⚖️ License

long-video-generation text-to-video

About

Tuning-Free Longer Video Diffusion via Noise Rescheduling

Example Output

Prompt:

"A chihuahua in astronaut suit floating in space, cinematic lighting, glow effect."

Output

Performance Metrics

795.08s Prediction Time

936.93s Total Time

All Input Parameters

{
  "prompt": "A chihuahua in astronaut suit floating in space, cinematic lighting, glow effect.",
  "save_fps": 10,
  "ddim_steps": 50,
  "num_frames": 32,
  "output_size": "576x1024",
  "window_size": 16,
  "window_stride": 4,
  "unconditional_guidance_scale": 12
}

Input Parameters

seed Type: integer: Random seed. Leave blank to randomize the seed
prompt Type: stringDefault: A chihuahua in astronaut suit floating in space, cinematic lighting, glow effect.: Prompt for video generation.
save_fps Type: integerDefault: 10: Frame per second for the generated video.
ddim_steps Type: integerDefault: 50: Number of denoising steps.
num_frames Type: integerDefault: 32: Number for frames to generate.
output_size Default: 576x1024: Choose the size of the output video.
window_size Type: integerDefault: 16: Window size.
window_stride Type: integerDefault: 4: Window stride.
unconditional_guidance_scale Type: numberDefault: 12: Classifier-free guidance scale.

Output Schema

Output

Type: string • Format: uri

Example Execution Logs

Using seed: 47023
Global seed set to 47023
DDIM scale True
ddim device cuda:0
/root/.pyenv/versions/3.11.4/lib/python3.11/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")

Version Details

Version ID: 15666961952fb74bec34968b89f0de59f15b9a31913dfade732053bda5f08f23
Version Created: October 31, 2023

Run on Replicate →