Towards Data Science By Anubhab Banerjee

How To Build Your Own LLM Runtime From Scratch

Read the original on Towards Data Science →

If you have ever wanted to actually build an LLM inference runtime yourself — pack your own weights, own every barrier, capture your own CUDA graphs — this is what that journey looks like on an H100. A step-by-step tour of a small runtime called annotated-llm-runtime, and the three bugs that produced most of the annotations.

Summary generated by The Flow from the publisher's feed. The full article lives at Towards Data Science.

arXiv AI
Jun 9

AgentCompile: An LLM-Guided Compiler for Direct CUDA Inference

arXiv:2606. 07665v1 Announce Type: cross Abstract: Transformer inference increasingly depends on specialized compiler and runtime support, but real model graphs still require semantic decisions about which regions are worth specializing and which CUDA implementation families are plausible.

By Xuanzhe Li, Ziyan Weng, Zhiyu Zhu, Junhui Hou
arXiv Machine Learning
Jun 11

MPK: A Compiler and Runtime for Mega-Kernelizing Tensor Programs

arXiv:2512. 22219v2 Announce Type: replace-cross Abstract: We introduce Mirage Persistent Kernel (MPK), the first compiler and runtime system that automatically transforms multi-GPU model inference into a single high-performance mega-kernel.

By Xinhao Cheng, Zhihao Zhang, Yu Zhou, Jianan Ji, Jinchen Jiang, Zepeng Zhao, Ziruo Xiao, Zihao Ye, Yingyi Huang, Ruihang Lai, Hongyi Jin, Bohan Hou, Mengdi Wu, Yixin Dong, Anthony Yip, Zihao Ye, Songting Wang, Wenqin Yang, Xupeng Miao, Tianqi Chen, Zhihao Jia