Alibaba Group - PAI

DAPPLE

An Efficient Pipelined Data Parallel Approach for Training Large Model

easyckpt

FlashModels

Fast and easy distributed model training examples.

FLASHNN

gradient-checkpointing

Make huge neural nets fit in memory

llumnix

Efficient and easy multi-instance LLM serving

Modelzoo-Data

DeepRec modelzoo's data set

one_shot_text_labeling

code and data for paper "One-shot Text Field Labeling using Attention and BeliefPropagation for Structure Information Extraction"

torchacc

PyTorch distributed training acceleration framework

xla

Enabling PyTorch on XLA Devices (e.g. Google TPU)

AlibabaPAI/llumnix

Efficient and easy multi-instance LLM serving

AlibabaPAI/FLASHNN

AlibabaPAI/DAPPLE

An Efficient Pipelined Data Parallel Approach for Training Large Model

AlibabaPAI/one_shot_text_labeling

code and data for paper "One-shot Text Field Labeling using Attention and BeliefPropagation for Structure Information Extraction"

AlibabaPAI/torchacc

PyTorch distributed training acceleration framework

AlibabaPAI/FlashModels

Fast and easy distributed model training examples.

AlibabaPAI/easyckpt

AlibabaPAI/gradient-checkpointing

Make huge neural nets fit in memory

AlibabaPAI/Modelzoo-Data

DeepRec modelzoo's data set

AlibabaPAI/xla

Enabling PyTorch on XLA Devices (e.g. Google TPU)

AlibabaPAI/EasyParallelLibrary

Easy Parallel Library (EPL) is a general and efficient deep learning framework for distributed giant model training.

AlibabaPAI/nni

An open source AutoML toolkit for automate machine learning lifecycle, including feature engineering, neural architecture search, model compression and hyper-parameter tuning.

AlibabaPAI/OLMo

Modeling, training, eval, and inference code for OLMo

AlibabaPAI/pytorch

Tensors and Dynamic neural networks in Python with strong GPU acceleration

AlibabaPAI/sparsehash-c11

Experimental C++11 version of sparsehash

AlibabaPAI/tpu-demos

AlibabaPAI/accelerate

🚀 A simple way to launch, train, and use PyTorch models on almost any device and distributed configuration, automatic mixed precision (including fp8), and easy-to-configure FSDP and DeepSpeed support

AlibabaPAI/FastChat_TorchAcc

An open platform for training, serving, and evaluating large language models. Release repo for Vicuna and Chatbot Arena.

AlibabaPAI/Megatron

Ongoing research training transformer models at scale

AlibabaPAI/openxla

A machine learning compiler for GPUs, CPUs, and ML accelerators

AlibabaPAI/reconfigurable-dl-scheduler

AlibabaPAI/transformers

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