LittleQili

transformers

🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

llm.c

LLM training in simple, raw C/CUDA

peft

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

LoRA

Code for loralib, an implementation of "LoRA: Low-Rank Adaptation of Large Language Models"

server

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

tiny-gpu

A minimal GPU design in Verilog to learn how GPUs work from the ground up

corenet

CoreNet: A library for training deep neural networks

miniforge

A conda-forge distribution.

lmdeploy

LMDeploy is a toolkit for compressing, deploying, and serving LLMs.

Liger-Kernel

Efficient Triton Kernels for LLM Training

ThunderKittens

Tile primitives for speedy kernels

gpushare-scheduler-extender

GPU Sharing Scheduler for Kubernetes Cluster

AzurePublicDataset

Microsoft Azure Traces

Overleaf-Workshop

Open Overleaf/ShareLaTex projects in vscode, with full collaboration support.

BitBLAS

BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.

DistServe

Disaggregated serving system for Large Language Models (LLMs).

rccl

ROCm Communication Collectives Library (RCCL)

varuna

flux

A fast communication-overlapping library for tensor parallelism on GPUs.

Caffeine

Caffeine for macOS 11+

TiledCUDA

TiledCUDA is a highly efficient kernel template library designed to elevate CUDA C’s level of abstraction for processing tiles.

SpotServe

SpotServe: Serving Generative Large Language Models on Preemptible Instances

orion

An interference-aware scheduler for fine-grained GPU sharing

paella

Paella: Low-latency Model Serving with Virtualized GPU Scheduling

pyjuice

Scalable training and inference for Probabilistic Circuits

rccl-tests

RCCL Performance Benchmark Tests

TiledKernel

TiledKernel is a code generation library based on macro kernels and memory hierarchy graph data structure.

amanda

chipgptft

Data is all you need: Finetuning LLMs for Chip Design via an Automated design-data augmentation framework (DAC 2024)

RAP-artifacts