Making large AI models cheaper, faster and more accessible
 
 
 
 
 
 
Go to file
Geng Zhang b6cc9313ef [NFC] polish colossalai/kernel/cuda_native/csrc/cpu_adam.cpp code style (#936) 2022-05-17 10:25:06 +08:00
.github [ci] added wheel build scripts (#910) 2022-05-05 16:06:39 +08:00
benchmark@607bb4a515 Automated submodule synchronization (#556) 2022-04-07 22:11:00 +08:00
colossalai [NFC] polish colossalai/kernel/cuda_native/csrc/cpu_adam.cpp code style (#936) 2022-05-17 10:25:06 +08:00
docker update setup and workflow (#222) 2022-02-15 11:31:13 +08:00
docs [refactor] moving memtracer to gemini (#801) 2022-04-19 10:13:08 +08:00
examples@b7be3b9076 Automated submodule synchronization (#960) 2022-05-14 21:55:34 +08:00
model_zoo [example] change qkv processing (#870) 2022-04-26 13:33:27 +08:00
requirements [CLI] refactored the launch CLI and fixed bugs in multi-node launching (#844) 2022-04-24 13:26:26 +08:00
tests [tensor] derive compute pattern from dist spec (#971) 2022-05-16 14:58:08 +08:00
.clang-format [tool] create .clang-format for pre-commit (#578) 2022-03-31 16:34:00 +08:00
.flake8 added flake8 config (#219) 2022-02-15 11:31:13 +08:00
.gitignore [model checkpoint] added unit tests for checkpoint save/load (#599) 2022-04-01 16:53:32 +08:00
.gitmodules fixed submodule url (#167) 2022-01-19 22:33:39 +08:00
.pre-commit-config.yaml [zero] find miss code (#378) 2022-03-11 15:50:28 +08:00
.readthedocs.yaml update doc requirements and rtd conf (#165) 2022-01-19 19:46:43 +08:00
.style.yapf fixed mkdir conflict and align yapf config with flake (#220) 2022-02-15 11:31:13 +08:00
CHANGE_LOG.md fix typo in CHANGE_LOG.md 2022-03-13 23:34:34 +09:00
CONTRIBUTING.md update contributing.md with the current workflow (#440) 2022-03-17 10:28:04 +08:00
LICENSE polish license (#300) 2022-03-11 15:50:28 +08:00
MANIFEST.in refactor kernel (#142) 2022-01-13 16:47:17 +08:00
README-zh-Hans.md update results on a single GPU, highlight quick view (#981) 2022-05-16 21:14:35 +08:00
README.md update results on a single GPU, highlight quick view (#981) 2022-05-16 21:14:35 +08:00
pytest.ini Migrated project 2021-10-28 18:21:23 +02:00
setup.py [setup] support more cuda architectures (#920) 2022-05-09 10:56:45 +08:00
version.txt udpate version (#982) 2022-05-17 09:48:14 +08:00

README.md

Colossal-AI

logo

An integrated large-scale model training system with efficient parallelization techniques.

Paper | Documentation | Examples | Forum | Blog

Build Documentation CodeFactor HuggingFace badge slack badge WeChat badge

| English | 中文 |

Table of Contents

Why Colossal-AI

Prof. James Demmel (UC Berkeley): Colossal-AI makes distributed training efficient, easy and scalable.

(back to top)

Features

Colossal-AI provides a collection of parallel training components for you. We aim to support you to write your distributed deep learning models just like how you write your model on your laptop. We provide user-friendly tools to kickstart distributed training in a few lines.

(back to top)

Parallel Demo

ViT

  • 14x larger batch size, and 5x faster training for Tensor Parallelism = 64

GPT-3

  • Save 50% GPU resources, and 10.7% acceleration

GPT-2

  • 11x lower GPU memory consumption, and superlinear scaling efficiency with Tensor Parallelism
  • 24x larger model size on the same hardware
  • over 3x acceleration

BERT

  • 2x faster training, or 50% longer sequence length

PaLM

Please visit our documentation and tutorials for more details.

(back to top)

Single GPU Demo

GPT-2

  • 20x larger model size on the same hardware

PaLM

  • 34x larger model size on the same hardware

(back to top)

Installation

Download From Official Releases

You can visit the Download page to download Colossal-AI with pre-built CUDA extensions.

Download From Source

The version of Colossal-AI will be in line with the main branch of the repository. Feel free to raise an issue if you encounter any problem. :)

git clone https://github.com/hpcaitech/ColossalAI.git
cd ColossalAI

# install dependency
pip install -r requirements/requirements.txt

# install colossalai
pip install .

If you don't want to install and enable CUDA kernel fusion (compulsory installation when using fused optimizer):

NO_CUDA_EXT=1 pip install .

(back to top)

Use Docker

Run the following command to build a docker image from Dockerfile provided.

cd ColossalAI
docker build -t colossalai ./docker

Run the following command to start the docker container in interactive mode.

docker run -ti --gpus all --rm --ipc=host colossalai bash

(back to top)

Community

Join the Colossal-AI community on Forum, Slack, and WeChat to share your suggestions, feedback, and questions with our engineering team.

Contributing

If you wish to contribute to this project, please follow the guideline in Contributing.

Thanks so much to all of our amazing contributors!

The order of contributor avatars is randomly shuffled.

(back to top)

Quick View

Start Distributed Training in Lines

parallel = dict(
    pipeline=2,
    tensor=dict(mode='2.5d', depth = 1, size=4)
)

Start Heterogeneous Training in Lines

zero = dict(
    model_config=dict(
        tensor_placement_policy='auto',
        shard_strategy=TensorShardStrategy(),
        reuse_fp16_shard=True
    ),
    optimizer_config=dict(initial_scale=2**5, gpu_margin_mem_ratio=0.2)
)

(back to top)

Cite Us

@article{bian2021colossal,
  title={Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training},
  author={Bian, Zhengda and Liu, Hongxin and Wang, Boxiang and Huang, Haichen and Li, Yongbin and Wang, Chuanrui and Cui, Fan and You, Yang},
  journal={arXiv preprint arXiv:2110.14883},
  year={2021}
}

(back to top)