yzh119

candle

Minimalist ML framework for Rust

server

The Triton Inference Server provides an optimized cloud and edge inferencing solution.

seL4

The seL4 microkernel

lorax

Multi-LoRA inference server that scales to 1000s of fine-tuned LLMs

docker-github-actions-runner

This will run the new self-hosted github actions runners with docker-in-docker

raft

RAFT contains fundamental widely-used algorithms and primitives for machine learning and information retrieval. The algorithms are CUDA-accelerated and form building blocks for more easily writing high performance applications.

LeanDojo

Tool for data extraction and interacting with Lean programmatically.

DCGM

NVIDIA Data Center GPU Manager (DCGM) is a project for gathering telemetry and measuring the health of NVIDIA GPUs

missed-optimizations

Missed optimizations in C compilers

jemallocator

Rust allocator using jemalloc as a backend

mega

Sequence modeling with Mega.

Atom

[MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving

Tensile

Stretching GPU performance for GEMMs and tensor contractions.

mscclpp

MSCCL++: A GPU-driven communication stack for scalable AI applications

wordflow

Social and customizable AI writing assistant! ✍️

fiddler

Fast Inference of MoE Models with CPU-GPU Orchestration

accelergy

Accelergy is an energy estimation infrastructure for accelerator energy estimations

tvm.tl

An extention of TVMScript to write simple and high performance GPU kernels with tensorcore.

lakeroad

FPGA synthesis tool powered by program synthesis

TransferBench

TransferBench is a utility capable of benchmarking simultaneous copies between user-specified devices (CPUs/GPUs)

mlc-assistant

Chat with your documents and improve your writing using large-language models within your browser.

PIM-DL-ASPLOS

PIM-DL: Expanding the Applicability of Commodity DRAM-PIMs for Deep Learning via Algorithm-System Co-Optimization

timeloop-python

Python wrapper for the timeloop project.

homework1

mlc-llm

Enable everyone to develop, optimize and deploy AI models natively on everyone's devices.

mscclpp-public

MSCCL++: A GPU-driven communication stack for scalable AI applications

punica-server-experimental

MindfulnessDepressionTherapy

佛法（限原始佛教）治疗抑郁症，有效抑郁症治疗方案。

stackage

Stable Haskell package sets: vetted consistent packages from Hackage

whl