Skip to yearly menu bar Skip to main content


Temporal Dynamic Quantization for Diffusion Models

Junhyuk So · Jungwon Lee · Daehyun Ahn · Hyungjun Kim · Eunhyeok Park

Great Hall & Hall B1+B2 (level 1) #611
[ ]
[ Paper [ Slides [ Poster [ OpenReview
Tue 12 Dec 8:45 a.m. PST — 10:45 a.m. PST


Diffusion model has gained popularity in vision applications due to its remarkable generative performance and versatility. However, its high storage and computation demands, resulting from the model size and iterative generation, hinder its use on mobile devices. Existing quantization techniques struggle to maintain performance even in 8-bit precision due to the diffusion model's unique property of temporal variation in activation. We introduce a novel quantization method that dynamically adjusts the quantization interval based on time step information, significantly improving output quality. Unlike conventional dynamic quantization techniques, our approach has no computational overhead during inference and is compatible with both post-training quantization (PTQ) and quantization-aware training (QAT). Our extensive experiments demonstrate substantial improvements in output quality with the quantized model across various configurations.

Chat is not available.