paper-with-me

Papers

MixDiT: Accelerating Image Diffusion Transformer Inference with Mixed-Precision MX Quantization

2025-04-11 · Daeun Kim, Jinwoo Hwang, Changhun Oh, Jongse Park

Diffusion Transformer (DiT) has driven significant progress in image generation tasks. However, DiT inferencing is notoriously compute-intensive and incurs long latency even on datacenter-scale GPUs, primarily due to its iterative nature and heavy reliance on GEMM operations inherent to its encoder-based structure. To address the challenge, prior work has explored quantization, but achieving low-precision quantization for DiT inferencing with both high accuracy and substantial speedup remains an open problem. To this end, this paper proposes MixDiT, an algorithm-hardware co-designed acceleration solution that exploits mixed Microscaling (MX) formats to quantize DiT activation values. MixDiT quantizes the DiT activation tensors by selectively applying higher precision to magnitude-based outliers, which produce mixed-precision GEMM operations. To achieve tangible speedup from the mixed-precision arithmetic, we design a MixDiT accelerator that enables precision-flexible multiplications and efficient MX precision conversions. Our experimental results show that MixDiT delivers a speedup of 2.10-5.32 times over RTX 3090, with no loss in FID.

📄 PDF Abstract BibTeX arXiv:2504.08398

Code (0)

등록된 구현이 없습니다.

Tasks

Image GenerationQuantization

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Residual Connection 설명 없음
Position-Wise Feed-Forward Layer 설명 없음
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Multi-Head Attention 설명 없음

Similar Papers 제목 키워드 기반

NAMI: Efficient Image Generation via Progressive Rectified Flow Transformers

2025-03-12 · Yuhang Ma, Bo Cheng, Shanyuan Liu, Ao Ma 외

Flow-based transformer models for image generation have achieved state-of-the-art performance with larger model parameters, but their inference deployment cost remains high. To enhance inference performance while maintai…

Image Generation

DuoDiff: Accelerating Diffusion Models with a Dual-Backbone Approach

2024-10-12 · Daniel Gallo Fernández, Răzvan-Andrei Matişan, Alejandro Monroy Muñoz, Ana-Maria Vasilcoiu 외

Diffusion models have achieved unprecedented performance in image generation, yet they suffer from slow inference due to their iterative sampling process. To address this, early-exiting has recently been proposed, where …

DenoisingImage Generation

Evolutionary Caching to Accelerate Your Off-the-Shelf Diffusion Model

2025-06-18 · Anirud Aggarwal, Abhinav Shrivastava, Matthew Gwilliam

Diffusion-based image generation models excel at producing high-quality synthetic content, but suffer from slow and computationally expensive inference. Prior work has attempted to mitigate this by caching and reusing fe…

Image Generation

SPADE: An Input-Adaptive Sparse Attention Engine for Fast Video Diffusion Models Inference

2026-08-04 · Shanghao Liu, Renze Chen, Size Zheng, Yuanqiang Liu 외 arxiv

Video diffusion transformers (vDiTs) generate high quality but pay quadratic self-attention cost, making inference prohibitive at video-token scales. The challenge is input-adaptive sparsity: selecting critical Q/K/V tok…

Video Generation

DiffSparse: Accelerating Diffusion Transformers with Learned Token Sparsity

2026-04-04 · Haowei Zhu, Ji Liu, Ziqiong Liu, Dong Li 외 arxiv

Diffusion models demonstrate outstanding performance in image generation, but their multi-step inference mechanism requires immense computational cost. Previous works accelerate inference by leveraging layer or token cac…

Image Generation