Saved in:
Bibliographic Details
Main Authors: Wang, Lei, Cheng, Yu, Shi, Yining, Tang, Zhengju, Mo, Zhiwen, Xie, Wenhao, Ma, Lingxiao, Xia, Yuqing, Xue, Jilong, Yang, Fan, Yang, Zhi
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.17577
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • Modern AI workloads rely heavily on optimized computing kernels for both training and inference. These AI kernels follow well-defined data-flow patterns, such as moving tiles between DRAM and SRAM and performing a sequence of computations on those tiles. However, writing high-performance kernels remains complex despite the clarity of these patterns. Achieving peak performance requires careful, hardware-centric optimizations to fully leverage modern accelerators. While domain-specific compilers attempt to reduce the burden of writing high-performance kernels, they often struggle with usability and expressiveness gaps. In this paper, we present TileLang, a generalized tiled programming model for more efficient AI Kernel programming. TileLang decouples scheduling space (thread binding, layout, tensorize and pipeline) from dataflow, and encapsulated them as a set of customization annotations and primitives. This approach allows users to focus on the kernel's data-flow itself, while leaving most other optimizations to compilers. We conduct comprehensive experiments on commonly-used devices, across numerous experiments, our evaluation shows that TileLang can achieve state-of-the-art performance in key kernels, demonstrating that its unified block-and-thread paradigm and transparent scheduling capabilities deliver both the power and flexibility demanded by modern AI system development.