Distributed Compiler based on Triton for Parallel Systems

Mirage Persistent Kernel: Compiling LLMs into a MegaKernel

kernels, of the mega variety

LightLLM is a Python-based LLM (Large Language Model) inference and serving framework, notable for its lightweight design, easy scalability, and high-speed performance.

📚 Modern C++ Tutorial: C++11/14/17/20 On the Fly | https://changkun.de/modern-cpp/

Timestep Embedding Tells: It's Time to Cache for Video Diffusion Model

Material for gpu-mode lectures

Yet Another Language Model: LLM inference in C++/CUDA, no libraries except for I/O

Nano vLLM

🙌 OpenHands: Code Less, Make More

A course of learning LLM inference serving on Apple Silicon for systems engineers: build a tiny vLLM + Qwen.

MiniCPM4 & MiniCPM4.1: Ultra-Efficient LLMs on End Devices, achieving 3+ generation speedup on reasoning tasks

MiniCPM-V 4.5: A GPT-4o Level MLLM for Single Image, Multi Image and High-FPS Video Understanding on Your Phone

Transforms complex documents like PDFs into LLM-ready markdown/JSON for your Agentic workflows.

Zplot demos

Source code for CSWAP-CLUSTER'21 and CSWAP+-TPDS'22

Source code for XPGraph-MICRO'22

Source code for iCache-HPCA'23

Appling the asynchronous tensor swapping to PyTorch framework.

Source code for GMBE-SC'23

Source code for CCLBTree-EuroSys'24

Source code for PHAST-TPDS'22

Source code for AdaMBE-SC'24

Source code for AMBEA-TC'24

Source code for ChunkGraph-ATC'24

Source code for CRouting