ColossalAI
Making large AI models cheaper, faster and more accessible
Stars: 38598
Colossal-AI is a deep learning system for large-scale parallel training. It provides a unified interface to scale sequential code of model training to distributed environments. Colossal-AI supports parallel training methods such as data, pipeline, tensor, and sequence parallelism and is integrated with heterogeneous training and zero redundancy optimizer.
README:
Colossal-AI: Making large AI models cheaper, faster, and more accessible
Paper | Documentation | Examples | Forum | GPU Cloud Playground | Blog
- [2024/06] Open-Sora Continues Open Source: Generate Any 16-Second 720p HD Video with One Click, Model Weights Ready to Use
- [2024/05] Large AI Models Inference Speed Doubled, Colossal-Inference Open Source Release
- [2024/04] Open-Sora Unveils Major Upgrade: Embracing Open Source with Single-Shot 16-Second Video Generation and 720p Resolution
- [2024/04] Most cost-effective solutions for inference, fine-tuning and pretraining, tailored to LLaMA3 series
- [2024/03] 314 Billion Parameter Grok-1 Inference Accelerated by 3.8x, Efficient and Easy-to-Use PyTorch+HuggingFace version is Here
- [2024/03] Open-Sora: Revealing Complete Model Parameters, Training Details, and Everything for Sora-like Video Generation Models
- [2024/03] Open-Sora:Sora Replication Solution with 46% Cost Reduction, Sequence Expansion to Nearly a Million
- [2024/01] Inference Performance Improved by 46%, Open Source Solution Breaks the Length Limit of LLM for Multi-Round Conversations
- [2023/07] HPC-AI Tech Raises 22 Million USD in Series A Funding
- Why Colossal-AI
- Features
-
Colossal-AI for Real World Applications
- Open-Sora: Revealing Complete Model Parameters, Training Details, and Everything for Sora-like Video Generation Models
- Colossal-LLaMA-2: One Half-Day of Training Using a Few Hundred Dollars Yields Similar Results to Mainstream Large Models, Open-Source and Commercial-Free Domain-Specific Llm Solution
- ColossalChat: An Open-Source Solution for Cloning ChatGPT With a Complete RLHF Pipeline
- AIGC: Acceleration of Stable Diffusion
- Biomedicine: Acceleration of AlphaFold Protein Structure
- Parallel Training Demo
- Single GPU Training Demo
- Inference
- Installation
- Use Docker
- Community
- Contributing
- Cite Us
Prof. James Demmel (UC Berkeley): Colossal-AI makes training AI models efficient, easy, and scalable.
Colossal-AI provides a collection of parallel 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 and inference in a few lines.
-
Parallelism strategies
- Data Parallelism
- Pipeline Parallelism
- 1D, 2D, 2.5D, 3D Tensor Parallelism
- Sequence Parallelism
- Zero Redundancy Optimizer (ZeRO)
- Auto-Parallelism
-
Heterogeneous Memory Management
-
Friendly Usage
- Parallelism based on the configuration file
Open-Sora:Revealing Complete Model Parameters, Training Details, and Everything for Sora-like Video Generation Models [code] [blog] [Model weights] [Demo] [GPU Cloud Playground] [OpenSora Image]
[GPU Cloud Playground] [LLaMA3 Image]
-
7B: One half-day of training using a few hundred dollars yields similar results to mainstream large models, open-source and commercial-free domain-specific LLM solution. [code] [blog] [HuggingFace model weights] [Modelscope model weights]
-
13B: Construct refined 13B private model with just $5000 USD. [code] [blog] [HuggingFace model weights] [Modelscope model weights]
Model | Backbone | Tokens Consumed | MMLU (5-shot) | CMMLU (5-shot) | AGIEval (5-shot) | GAOKAO (0-shot) | CEval (5-shot) |
---|---|---|---|---|---|---|---|
Baichuan-7B | - | 1.2T | 42.32 (42.30) | 44.53 (44.02) | 38.72 | 36.74 | 42.80 |
Baichuan-13B-Base | - | 1.4T | 50.51 (51.60) | 55.73 (55.30) | 47.20 | 51.41 | 53.60 |
Baichuan2-7B-Base | - | 2.6T | 46.97 (54.16) | 57.67 (57.07) | 45.76 | 52.60 | 54.00 |
Baichuan2-13B-Base | - | 2.6T | 54.84 (59.17) | 62.62 (61.97) | 52.08 | 58.25 | 58.10 |
ChatGLM-6B | - | 1.0T | 39.67 (40.63) | 41.17 (-) | 40.10 | 36.53 | 38.90 |
ChatGLM2-6B | - | 1.4T | 44.74 (45.46) | 49.40 (-) | 46.36 | 45.49 | 51.70 |
InternLM-7B | - | 1.6T | 46.70 (51.00) | 52.00 (-) | 44.77 | 61.64 | 52.80 |
Qwen-7B | - | 2.2T | 54.29 (56.70) | 56.03 (58.80) | 52.47 | 56.42 | 59.60 |
Llama-2-7B | - | 2.0T | 44.47 (45.30) | 32.97 (-) | 32.60 | 25.46 | - |
Linly-AI/Chinese-LLaMA-2-7B-hf | Llama-2-7B | 1.0T | 37.43 | 29.92 | 32.00 | 27.57 | - |
wenge-research/yayi-7b-llama2 | Llama-2-7B | - | 38.56 | 31.52 | 30.99 | 25.95 | - |
ziqingyang/chinese-llama-2-7b | Llama-2-7B | - | 33.86 | 34.69 | 34.52 | 25.18 | 34.2 |
TigerResearch/tigerbot-7b-base | Llama-2-7B | 0.3T | 43.73 | 42.04 | 37.64 | 30.61 | - |
LinkSoul/Chinese-Llama-2-7b | Llama-2-7B | - | 48.41 | 38.31 | 38.45 | 27.72 | - |
FlagAlpha/Atom-7B | Llama-2-7B | 0.1T | 49.96 | 41.10 | 39.83 | 33.00 | - |
IDEA-CCNL/Ziya-LLaMA-13B-v1.1 | Llama-13B | 0.11T | 50.25 | 40.99 | 40.04 | 30.54 | - |
Colossal-LLaMA-2-7b-base | Llama-2-7B | 0.0085T | 53.06 | 49.89 | 51.48 | 58.82 | 50.2 |
Colossal-LLaMA-2-13b-base | Llama-2-13B | 0.025T | 56.42 | 61.80 | 54.69 | 69.53 | 60.3 |
ColossalChat: An open-source solution for cloning ChatGPT with a complete RLHF pipeline. [code] [blog] [demo] [tutorial]
- Up to 10 times faster for RLHF PPO Stage3 Training
- Up to 7.73 times faster for single server training and 1.42 times faster for single-GPU inference
- Up to 10.3x growth in model capacity on one GPU
- A mini demo training process requires only 1.62GB of GPU memory (any consumer-grade GPU)
- Increase the capacity of the fine-tuning model by up to 3.7 times on a single GPU
- Keep at a sufficiently high running speed
Acceleration of AIGC (AI-Generated Content) models such as Stable Diffusion v1 and Stable Diffusion v2.
- Training: Reduce Stable Diffusion memory consumption by up to 5.6x and hardware cost by up to 46x (from A100 to RTX3060).
- DreamBooth Fine-tuning: Personalize your model using just 3-5 images of the desired subject.
- Inference: Reduce inference GPU memory consumption by 2.5x.
Acceleration of AlphaFold Protein Structure
- FastFold: Accelerating training and inference on GPU Clusters, faster data processing, inference sequence containing more than 10000 residues.
- FastFold with Intel: 3x inference acceleration and 39% cost reduce.
- xTrimoMultimer: accelerating structure prediction of protein monomers and multimer by 11x.
- 70 billion parameter LLaMA3 model training accelerated by 18% [code] [GPU Cloud Playground] [LLaMA3 Image]
- Enhanced MoE parallelism, Open-source MoE model training can be 9 times more efficient [code] [blog]
- Save 50% GPU resources and 10.7% acceleration
- 11x lower GPU memory consumption, and superlinear scaling efficiency with Tensor Parallelism
- 24x larger model size on the same hardware
- over 3x acceleration
- 2x faster training, or 50% longer sequence length
- PaLM-colossalai: Scalable implementation of Google's Pathways Language Model (PaLM).
- Open Pretrained Transformer (OPT), a 175-Billion parameter AI language model released by Meta, which stimulates AI programmers to perform various downstream tasks and application deployments because of public pre-trained model weights.
- 45% speedup fine-tuning OPT at low cost in lines. [Example] [Online Serving]
Please visit our documentation and examples for more details.
- 14x larger batch size, and 5x faster training for Tensor Parallelism = 64
- Cached Embedding, utilize software cache to train larger embedding tables with a smaller GPU memory budget.
- 20x larger model size on the same hardware
- 120x larger model size on the same hardware (RTX 3080)
- 34x larger model size on the same hardware
- Large AI models inference speed doubled, compared to the offline inference performance of vLLM in some cases. [code] [blog] [GPU Cloud Playground] [LLaMA3 Image]
- 314 Billion Parameter Grok-1 Inference Accelerated by 3.8x, an easy-to-use Python + PyTorch + HuggingFace version for Inference.
[code] [blog] [HuggingFace Grok-1 PyTorch model weights] [ModelScope Grok-1 PyTorch model weights]
- SwiftInfer: Inference performance improved by 46%, open source solution breaks the length limit of LLM for multi-round conversations
Requirements:
- PyTorch >= 2.2
- Python >= 3.7
- CUDA >= 11.0
- NVIDIA GPU Compute Capability >= 7.0 (V100/RTX20 and higher)
- Linux OS
If you encounter any problem with installation, you may want to raise an issue in this repository.
You can easily install Colossal-AI with the following command. By default, we do not build PyTorch extensions during installation.
pip install colossalai
Note: only Linux is supported for now.
However, if you want to build the PyTorch extensions during installation, you can set BUILD_EXT=1
.
BUILD_EXT=1 pip install colossalai
Otherwise, CUDA kernels will be built during runtime when you actually need them.
We also keep releasing the nightly version to PyPI every week. This allows you to access the unreleased features and bug fixes in the main branch. Installation can be made via
pip install colossalai-nightly
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 problems. :)
git clone https://github.com/hpcaitech/ColossalAI.git
cd ColossalAI
# install colossalai
pip install .
By default, we do not compile CUDA/C++ kernels. ColossalAI will build them during runtime. If you want to install and enable CUDA kernel fusion (compulsory installation when using fused optimizer):
BUILD_EXT=1 pip install .
For Users with CUDA 10.2, you can still build ColossalAI from source. However, you need to manually download the cub library and copy it to the corresponding directory.
# clone the repository
git clone https://github.com/hpcaitech/ColossalAI.git
cd ColossalAI
# download the cub library
wget https://github.com/NVIDIA/cub/archive/refs/tags/1.8.0.zip
unzip 1.8.0.zip
cp -r cub-1.8.0/cub/ colossalai/kernel/cuda_native/csrc/kernels/include/
# install
BUILD_EXT=1 pip install .
You can directly pull the docker image from our DockerHub page. The image is automatically uploaded upon release.
Run the following command to build a docker image from Dockerfile provided.
Building Colossal-AI from scratch requires GPU support, you need to use Nvidia Docker Runtime as the default when doing
docker build
. More details can be found here. We recommend you install Colossal-AI from our project page directly.
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
Join the Colossal-AI community on Forum, Slack, and WeChat(微信) to share your suggestions, feedback, and questions with our engineering team.
Referring to the successful attempts of BLOOM and Stable Diffusion, any and all developers and partners with computing powers, datasets, models are welcome to join and build the Colossal-AI community, making efforts towards the era of big AI models!
You may contact us or participate in the following ways:
- Leaving a Star ⭐ to show your like and support. Thanks!
- Posting an issue, or submitting a PR on GitHub follow the guideline in Contributing
- Send your official proposal to email [email protected]
Thanks so much to all of our amazing contributors!
We leverage the power of GitHub Actions to automate our development, release and deployment workflows. Please check out this documentation on how the automated workflows are operated.
This project is inspired by some related projects (some by our team and some by other organizations). We would like to credit these amazing projects as listed in the Reference List.
To cite this project, you can use the following BibTeX citation.
@inproceedings{10.1145/3605573.3605613,
author = {Li, Shenggui and Liu, Hongxin and Bian, Zhengda and Fang, Jiarui and Huang, Haichen and Liu, Yuliang and Wang, Boxiang and You, Yang},
title = {Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training},
year = {2023},
isbn = {9798400708435},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3605573.3605613},
doi = {10.1145/3605573.3605613},
abstract = {The success of Transformer models has pushed the deep learning model scale to billions of parameters, but the memory limitation of a single GPU has led to an urgent need for training on multi-GPU clusters. However, the best practice for choosing the optimal parallel strategy is still lacking, as it requires domain expertise in both deep learning and parallel computing. The Colossal-AI system addressed the above challenge by introducing a unified interface to scale your sequential code of model training to distributed environments. It supports parallel training methods such as data, pipeline, tensor, and sequence parallelism and is integrated with heterogeneous training and zero redundancy optimizer. Compared to the baseline system, Colossal-AI can achieve up to 2.76 times training speedup on large-scale models.},
booktitle = {Proceedings of the 52nd International Conference on Parallel Processing},
pages = {766–775},
numpages = {10},
keywords = {datasets, gaze detection, text tagging, neural networks},
location = {Salt Lake City, UT, USA},
series = {ICPP '23}
}
Colossal-AI has been accepted as official tutorial by top conferences NeurIPS, SC, AAAI, PPoPP, CVPR, ISC, NVIDIA GTC ,etc.
For Tasks:
Click tags to check more tools for each tasksFor Jobs:
Alternative AI tools for ColossalAI
Similar Open Source Tools
ColossalAI
Colossal-AI is a deep learning system for large-scale parallel training. It provides a unified interface to scale sequential code of model training to distributed environments. Colossal-AI supports parallel training methods such as data, pipeline, tensor, and sequence parallelism and is integrated with heterogeneous training and zero redundancy optimizer.
TokenPacker
TokenPacker is a novel visual projector that compresses visual tokens by 75%∼89% with high efficiency. It adopts a 'coarse-to-fine' scheme to generate condensed visual tokens, achieving comparable or better performance across diverse benchmarks. The tool includes TokenPacker for general use and TokenPacker-HD for high-resolution image understanding. It provides training scripts, checkpoints, and supports various compression ratios and patch numbers.
aim
Aim is an open-source, self-hosted ML experiment tracking tool designed to handle 10,000s of training runs. Aim provides a performant and beautiful UI for exploring and comparing training runs. Additionally, its SDK enables programmatic access to tracked metadata — perfect for automations and Jupyter Notebook analysis. **Aim's mission is to democratize AI dev tools 🎯**
nuitrack-sdk
Nuitrack™ is an ultimate 3D body tracking solution developed by 3DiVi Inc. It enables body motion analytics applications for virtually any widespread depth sensors and hardware platforms, supporting a wide range of applications from real-time gesture recognition on embedded platforms to large-scale multisensor analytical systems. Nuitrack provides highly-sophisticated 3D skeletal tracking, basic facial analysis, hand tracking, and gesture recognition APIs for UI control. It offers two skeletal tracking engines: classical for embedded hardware and AI for complex poses, providing a human-centric spatial understanding tool for natural and intelligent user engagement.
AI-Notes
AI-Notes is a repository dedicated to practical applications of artificial intelligence and deep learning. It covers concepts such as data mining, machine learning, natural language processing, and AI. The repository contains Jupyter Notebook examples for hands-on learning and experimentation. It explores the development stages of AI, from narrow artificial intelligence to general artificial intelligence and superintelligence. The content delves into machine learning algorithms, deep learning techniques, and the impact of AI on various industries like autonomous driving and healthcare. The repository aims to provide a comprehensive understanding of AI technologies and their real-world applications.
Embodied-AI-Guide
Embodied-AI-Guide is a comprehensive guide for beginners to understand Embodied AI, focusing on the path of entry and useful information in the field. It covers topics such as Reinforcement Learning, Imitation Learning, Large Language Model for Robotics, 3D Vision, Control, Benchmarks, and provides resources for building cognitive understanding. The repository aims to help newcomers quickly establish knowledge in the field of Embodied AI.
painting-droid
Painting Droid is an AI-powered cross-platform painting app inspired by MS Paint, expandable with plugins and open. It utilizes various AI models, from paid providers to self-hosted open-source models, as well as some lightweight ones built into the app. Features include regular painting app features, AI-generated content filling and augmentation, filters and effects, image manipulation, plugin support, and cross-platform compatibility.
big-AGI
big-AGI is an AI suite designed for professionals seeking function, form, simplicity, and speed. It offers best-in-class Chats, Beams, and Calls with AI personas, visualizations, coding, drawing, side-by-side chatting, and more, all wrapped in a polished UX. The tool is powered by the latest models from 12 vendors and open-source servers, providing users with advanced AI capabilities and a seamless user experience. With continuous updates and enhancements, big-AGI aims to stay ahead of the curve in the AI landscape, catering to the needs of both developers and AI enthusiasts.
audio-webui
Audio Webui is a tool designed to provide a user-friendly interface for audio processing tasks. It supports automatic installers, Docker deployment, local manual installation, Google Colab integration, and common command line flags. Users can easily download, install, update, and run the tool for various audio-related tasks. The tool requires Python 3.10, Git, and ffmpeg for certain features. It also offers extensions for additional functionalities.
GIMP-ML
A.I. for GNU Image Manipulation Program (GIMP-ML) is a repository that provides Python plugins for using computer vision models in GIMP. The code base and models are continuously updated to support newer and more stable functionality. Users can edit images with text, outpaint images, and generate images from text using models like Dalle 2 and Dalle 3. The repository encourages citations using a specific bibtex entry and follows the MIT license for GIMP-ML and the original models.
agentscope
AgentScope is a multi-agent platform designed to empower developers to build multi-agent applications with large-scale models. It features three high-level capabilities: Easy-to-Use, High Robustness, and Actor-Based Distribution. AgentScope provides a list of `ModelWrapper` to support both local model services and third-party model APIs, including OpenAI API, DashScope API, Gemini API, and ollama. It also enables developers to rapidly deploy local model services using libraries such as ollama (CPU inference), Flask + Transformers, Flask + ModelScope, FastChat, and vllm. AgentScope supports various services, including Web Search, Data Query, Retrieval, Code Execution, File Operation, and Text Processing. Example applications include Conversation, Game, and Distribution. AgentScope is released under Apache License 2.0 and welcomes contributions.
BitBLAS
BitBLAS is a library for mixed-precision BLAS operations on GPUs, for example, the $W_{wdtype}A_{adtype}$ mixed-precision matrix multiplication where $C_{cdtype}[M, N] = A_{adtype}[M, K] \times W_{wdtype}[N, K]$. BitBLAS aims to support efficient mixed-precision DNN model deployment, especially the $W_{wdtype}A_{adtype}$ quantization in large language models (LLMs), for example, the $W_{UINT4}A_{FP16}$ in GPTQ, the $W_{INT2}A_{FP16}$ in BitDistiller, the $W_{INT2}A_{INT8}$ in BitNet-b1.58. BitBLAS is based on techniques from our accepted submission at OSDI'24.
intel-extension-for-tensorflow
Intel® Extension for TensorFlow* is a high performance deep learning extension plugin based on TensorFlow PluggableDevice interface. It aims to accelerate AI workloads by allowing users to plug Intel CPU or GPU devices into TensorFlow on-demand, exposing the computing power inside Intel's hardware. The extension provides XPU specific implementation, kernels & operators, graph optimizer, device runtime, XPU configuration management, XPU backend selection, and options for turning on/off advanced features.
Q-Bench
Q-Bench is a benchmark for general-purpose foundation models on low-level vision, focusing on multi-modality LLMs performance. It includes three realms for low-level vision: perception, description, and assessment. The benchmark datasets LLVisionQA and LLDescribe are collected for perception and description tasks, with open submission-based evaluation. An abstract evaluation code is provided for assessment using public datasets. The tool can be used with the datasets API for single images and image pairs, allowing for automatic download and usage. Various tasks and evaluations are available for testing MLLMs on low-level vision tasks.
pro-chat
ProChat is a components library focused on quickly building large language model chat interfaces. It empowers developers to create rich, dynamic, and intuitive chat interfaces with features like automatic chat caching, streamlined conversations, message editing tools, auto-rendered Markdown, and programmatic controls. The tool also includes design evolution plans such as customized dialogue rendering, enhanced request parameters, personalized error handling, expanded documentation, and atomic component design.
For similar tasks
ColossalAI
Colossal-AI is a deep learning system for large-scale parallel training. It provides a unified interface to scale sequential code of model training to distributed environments. Colossal-AI supports parallel training methods such as data, pipeline, tensor, and sequence parallelism and is integrated with heterogeneous training and zero redundancy optimizer.
For similar jobs
weave
Weave is a toolkit for developing Generative AI applications, built by Weights & Biases. With Weave, you can log and debug language model inputs, outputs, and traces; build rigorous, apples-to-apples evaluations for language model use cases; and organize all the information generated across the LLM workflow, from experimentation to evaluations to production. Weave aims to bring rigor, best-practices, and composability to the inherently experimental process of developing Generative AI software, without introducing cognitive overhead.
agentcloud
AgentCloud is an open-source platform that enables companies to build and deploy private LLM chat apps, empowering teams to securely interact with their data. It comprises three main components: Agent Backend, Webapp, and Vector Proxy. To run this project locally, clone the repository, install Docker, and start the services. The project is licensed under the GNU Affero General Public License, version 3 only. Contributions and feedback are welcome from the community.
oss-fuzz-gen
This framework generates fuzz targets for real-world `C`/`C++` projects with various Large Language Models (LLM) and benchmarks them via the `OSS-Fuzz` platform. It manages to successfully leverage LLMs to generate valid fuzz targets (which generate non-zero coverage increase) for 160 C/C++ projects. The maximum line coverage increase is 29% from the existing human-written targets.
LLMStack
LLMStack is a no-code platform for building generative AI agents, workflows, and chatbots. It allows users to connect their own data, internal tools, and GPT-powered models without any coding experience. LLMStack can be deployed to the cloud or on-premise and can be accessed via HTTP API or triggered from Slack or Discord.
VisionCraft
The VisionCraft API is a free API for using over 100 different AI models. From images to sound.
kaito
Kaito is an operator that automates the AI/ML inference model deployment in a Kubernetes cluster. It manages large model files using container images, avoids tuning deployment parameters to fit GPU hardware by providing preset configurations, auto-provisions GPU nodes based on model requirements, and hosts large model images in the public Microsoft Container Registry (MCR) if the license allows. Using Kaito, the workflow of onboarding large AI inference models in Kubernetes is largely simplified.
PyRIT
PyRIT is an open access automation framework designed to empower security professionals and ML engineers to red team foundation models and their applications. It automates AI Red Teaming tasks to allow operators to focus on more complicated and time-consuming tasks and can also identify security harms such as misuse (e.g., malware generation, jailbreaking), and privacy harms (e.g., identity theft). The goal is to allow researchers to have a baseline of how well their model and entire inference pipeline is doing against different harm categories and to be able to compare that baseline to future iterations of their model. This allows them to have empirical data on how well their model is doing today, and detect any degradation of performance based on future improvements.
Azure-Analytics-and-AI-Engagement
The Azure-Analytics-and-AI-Engagement repository provides packaged Industry Scenario DREAM Demos with ARM templates (Containing a demo web application, Power BI reports, Synapse resources, AML Notebooks etc.) that can be deployed in a customer’s subscription using the CAPE tool within a matter of few hours. Partners can also deploy DREAM Demos in their own subscriptions using DPoC.