LLM inference in C/C++

A high-throughput and memory-efficient inference and serving engine for LLMs

A post-modern modal text editor.

程序员延寿指南 | A programmer's guide to live longer

Everything you need to move your project faster

AISystem 主要是指AI系统，包括AI芯片、AI编译器、AI推理和训练框架等AI全栈底层技术

CUDA on non-NVIDIA GPUs

本人的科研经验

📚LeetCUDA: Modern CUDA Learn Notes with PyTorch for Beginners🐑, 200+ CUDA Kernels, Tensor Cores, HGEMM, FA-2 MMA.🎉

Material for gpu-mode lectures

😏国内外计算机的优秀课程，包含MIT、CMU等世界CS名校，🔥🔥其中包含计算机基础学科（操作系统、计算机网络、编译器、数据库、数据结构与算法等）以及人工智能&AI等高级科目，欢迎通过PR形式贡献！

[TMLR 2024] Efficient Large Language Models: A Survey

This is a series of GPU optimization topics. Here we will introduce how to optimize the CUDA kernel in detail. I will introduce several basic kernel optimizations, including: elementwise, reduce, s…

My curriculum vitae (CV) written using LaTeX.

List of papers related to neural network quantization in recent AI conferences and journals.

hpc-learning

欧港新CS留学项目指北

[NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantization

[ICML 2024] KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache

[ICML 2024] BiLLM: Pushing the Limit of Post-Training Quantization for LLMs

A parallel programming training mini app simulating weather-like flows

Wiki fo HPC

BGHT: High-performance static GPU hash tables.

zTT: Learning-based DVFS with Zero Thermal Throttling for Mobile Devices [MobiSys'21] - Artifact Evaluation

Scalable radix top-k selection on GPUs.

a Model-Free GPU Online Energy Optimization (MF-GPOEO) framework

A quick survival guild for i18n students who comes to chalmers.

XiTAO is a lightweight layer built on top of modern C++ features with the goals of being low-overhead and serving as a development platform for testing scheduling and resource management algorithms.