OpenTrain AI
Maintained implementation availablepytorchPretrained Models Available

Score identity Distillation: Exponentially Fast Distillation of Pretrained Diffusion Models for One-Step Generation

April 1, 2024arXiv: 2404.04057
1 repo151 stars~a few hours to reproduce
arXiv PDF

Abstract

Best Implementation

PyTorch code and model checkpoints for Score identity Distillation (SiD) and its adversarial version (SiDA)

151 10 Mar 2025 Apache-2.0
License
CI
Deps
Docker
  • Selected mingyuanzhou/sid as the strongest maintained implementation for new work.
  • Includes dependency/environment manifest signals.
  • Repository activity is within the last 24 months.

Reproduction Path

  1. 1

    Start with mingyuanzhou/sid and validate setup instructions in README.

  2. 2

    Reproduce the baseline result with the provided defaults before modifying hyperparameters.

  3. 3

    Log exact dependency versions and runtime environment for reproducibility.

Time to first repro: a few hoursNo CI workflows detected

Additional Implementations

No additional verified repositories beyond the primary recommendation.

Hugging Face Artifacts

No direct paper-linked artifacts were found. Showing strongest curated related artifacts.

Research Context

Methods