PQD: Post-training Quantization for Efficient Diffusion Models
This work addresses efficiency issues for users of diffusion models in image generation, but it is incremental as it builds on existing quantization techniques.
The paper tackles the high computational cost and slow generation speed of diffusion models by proposing a post-training quantization method (PQD) that reduces model size to 8-bit or 4-bit with minimal performance loss, achieving only a few FID change on ImageNet.
Diffusionmodels(DMs)havedemonstratedremarkableachievements in synthesizing images of high fidelity and diversity. However, the extensive computational requirements and slow generative speed of diffusion models have limited their widespread adoption. In this paper, we propose a novel post-training quantization for diffusion models (PQD), which is a time-aware optimization framework for diffusion models based on post-training quantization. The proposed framework optimizes the inference process by selecting representative samples and conducting time-aware calibration. Experimental results show that our proposed method is able to directly quantize full-precision diffusion models into 8-bit or 4-bit models while maintaining comparable performance in a training-free manner, achieving a few FID change on ImageNet for unconditional image generation. Our approach demonstrates compatibility and can also be applied to 512x512 text-guided image generation for the first time.