zhan4808/cache-barrier
exploring the L2 cache barrier in MLA reconstruction GEMMs and why INT4 quantization fails to reduce decode latency
Repositories
zhan4808/kernel-compass
hardware counter-grounded kernel optimization for transformer GEMMs
zhan4808/zhan4808
Config files for my GitHub profile.
zhan4808/chatui
zhan4808/FlagGems
FlagGems is an operator library for large language models implemented in the Triton Language.
zhan4808/robert
personal portfolio website
zhan4808/TSL
CW&T Time Since Launch
zhan4808/systems-sw-report
zhan4808/tiny-gemm
decode-focused triton fused int4 gemm and transformer kernels for small-batch LLM inference
zhan4808/sglang
SGLang is a high-performance serving framework for large language models and multimodal models.
zhan4808/vllm
A high-throughput and memory-efficient inference and serving engine for LLMs
zhan4808/MLSys
Hosts MLSys Competition Track Problems
zhan4808/modded-nanogpt
NanoGPT (124M) in 2 minutes
zhan4808/ece60850-m1
zhan4808/flashinfer2026
FlashInfer Bench @ MLSys 2026: Building AI agents to write high performance GPU kernels
zhan4808/usrintent
zhan4808/ThunderKittens
Tile primitives for speedy kernels
zhan4808/llvm-project
The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.
zhan4808/openclaw
Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞
zhan4808/pytorchdemo
zhan4808/ece570proj
zhan4808/GRDB.swift
A toolkit for SQLite databases, with a focus on application development
zhan4808/slynk
real time videos you can talk to
zhan4808/socialmesh
building meaningful relationships with intelligent mesh and automated networking
zhan4808/tensorboard
TensorFlow's Visualization Toolkit
zhan4808/boilermake2025
zhan4808/labs
Public Lab Documents for ECE 36200
zhan4808/artsage
your own museum curator
zhan4808/triton
Development repository for the Triton language and compiler
zhan4808/gptq
GPTQ: Accurate Post-training Quantization of Generative Pretrained Transformers