FastLightGen: Fast and Light Video Generation with Fewer Steps and Parameters

TL;DR

FastLightGen introduces a novel method to create lightweight, fast video generation models by jointly compressing model size and reducing inference steps, significantly improving efficiency without sacrificing quality.

Contribution

The paper presents FastLightGen, a new algorithm that optimally distills large video generation models into smaller, faster versions by jointly compressing parameters and inference steps.

Findings

Achieves high visual quality with 4-step sampling and 30% parameter pruning.

Outperforms existing methods in efficiency and quality on benchmark datasets.

Establishes a new state-of-the-art in fast, lightweight video generation.

Abstract

The recent advent of powerful video generation models, such as Hunyuan, WanX, Veo3, and Kling, has inaugurated a new era in the field. However, the practical deployment of these models is severely impeded by their substantial computational overhead, which stems from enormous parameter counts and the iterative, multi-step sampling process required during inference. Prior research on accelerating generative models has predominantly followed two distinct trajectories: reducing the number of sampling steps (e.g., LCM, DMD, and MagicDistillation) or compressing the model size for more efficient inference (e.g., ICMD). The potential of simultaneously compressing both to create a fast and lightweight model remains an unexplored avenue. In this paper, we propose FastLightGen, an algorithm that transforms large, computationally expensive models into fast, lightweight counterparts. The core idea is to construct an optimal teacher model, one engineered to maximize student performance, within a synergistic framework for distilling both model size and inference steps. Our extensive experiments on HunyuanVideo-ATI2V and WanX-TI2V reveal that a generator using 4-step sampling and 30\% parameter pruning achieves optimal visual quality under a constrained inference budget. Furthermore, FastLightGen consistently outperforms all competing methods, establishing a new state-of-the-art in efficient video generation.