*[2023/09] [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](https://www.hpc-ai.tech/blog/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)
*[2024/01] [Inference Performance Improved by 46%, Open Source Solution Breaks the Length Limit of LLM for Multi-Round Conversations](https://hpc-ai.com/blog/Colossal-AI-SwiftInfer)
*[2024/01] [Construct Refined 13B Private Model With Just $5000 USD, Upgraded Colossal-AI Llama-2 Open Source](https://hpc-ai.com/blog/colossal-llama-2-13b)
*[2023/11] [Enhanced MoE Parallelism, Open-source MoE Model Training Can Be 9 Times More Efficient](https://www.hpc-ai.tech/blog/enhanced-moe-parallelism-open-source-moe-model-training-can-be-9-times-more-efficient)
*[2023/09] [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](https://www.hpc-ai.tech/blog/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)
*[2023/09] [70 Billion Parameter LLaMA2 Model Training Accelerated by 195%](https://www.hpc-ai.tech/blog/70b-llama2-training)
*[2023/09] [70 Billion Parameter LLaMA2 Model Training Accelerated by 195%](https://www.hpc-ai.tech/blog/70b-llama2-training)
*[2023/07] [HPC-AI Tech Raises 22 Million USD in Series A Funding](https://www.hpc-ai.tech/blog/hpc-ai-tech-raises-22-million-usd-in-series-a-funding-to-fuel-team-expansion-and-business-growth)
*[2023/07] [HPC-AI Tech Raises 22 Million USD in Series A Funding](https://www.hpc-ai.tech/blog/hpc-ai-tech-raises-22-million-usd-in-series-a-funding-to-fuel-team-expansion-and-business-growth)
*[2023/07] [65B Model Pretraining Accelerated by 38%, Best Practices for Building LLaMA-Like Base Models Open-Source](https://www.hpc-ai.tech/blog/large-model-pretraining)
*[2023/07] [65B Model Pretraining Accelerated by 38%, Best Practices for Building LLaMA-Like Base Models Open-Source](https://www.hpc-ai.tech/blog/large-model-pretraining)
*[2023/03] [ColossalChat: An Open-Source Solution for Cloning ChatGPT With a Complete RLHF Pipeline](https://medium.com/@yangyou_berkeley/colossalchat-an-open-source-solution-for-cloning-chatgpt-with-a-complete-rlhf-pipeline-5edf08fb538b)
*[2023/03] [ColossalChat: An Open-Source Solution for Cloning ChatGPT With a Complete RLHF Pipeline](https://medium.com/@yangyou_berkeley/colossalchat-an-open-source-solution-for-cloning-chatgpt-with-a-complete-rlhf-pipeline-5edf08fb538b)
*[2023/03] [Intel and Colossal-AI Partner to Deliver Cost-Efficient Open-Source Solution for Protein Folding Structure Prediction](https://www.hpc-ai.tech/blog/intel-habana)
*[2023/03] [Intel and Colossal-AI Partner to Deliver Cost-Efficient Open-Source Solution for Protein Folding Structure Prediction](https://www.hpc-ai.tech/blog/intel-habana)
*[2023/03] [AWS and Google Fund Colossal-AI with Startup Cloud Programs](https://www.hpc-ai.tech/blog/aws-and-google-fund-colossal-ai-with-startup-cloud-programs)
*[2023/03] [AWS and Google Fund Colossal-AI with Startup Cloud Programs](https://www.hpc-ai.tech/blog/aws-and-google-fund-colossal-ai-with-startup-cloud-programs)
*[2023/02] [Open Source Solution Replicates ChatGPT Training Process! Ready to go with only 1.6GB GPU Memory](https://www.hpc-ai.tech/blog/colossal-ai-chatgpt)
*[2023/01] [Hardware Savings Up to 46 Times for AIGC and Automatic Parallelism](https://medium.com/pytorch/latest-colossal-ai-boasts-novel-automatic-parallelism-and-offers-savings-up-to-46x-for-stable-1453b48f3f02)
## Table of Contents
## Table of Contents
<ul>
<ul>
...
@@ -52,6 +53,7 @@
...
@@ -52,6 +53,7 @@
<ahref="#Parallel-Training-Demo">Parallel Training Demo</a>
<ahref="#Parallel-Training-Demo">Parallel Training Demo</a>
<palign="right">(<ahref="#top">back to top</a>)</p>
<palign="right">(<ahref="#top">back to top</a>)</p>
## Colossal-AI in the Real World
## Colossal-AI in the Real World
### Colossal-LLaMA-2
### 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.
-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.
@@ -215,7 +219,7 @@ Acceleration of AIGC (AI-Generated Content) models such as [Stable Diffusion v1]
...
@@ -215,7 +219,7 @@ Acceleration of AIGC (AI-Generated Content) models such as [Stable Diffusion v1]
-[DreamBooth Fine-tuning](https://github.com/hpcaitech/ColossalAI/tree/main/examples/images/dreambooth): Personalize your model using just 3-5 images of the desired subject.
-[DreamBooth Fine-tuning](https://github.com/hpcaitech/ColossalAI/tree/main/examples/images/dreambooth): Personalize your model using just 3-5 images of the desired subject.
-[SwiftInfer](https://github.com/hpcaitech/SwiftInfer): Inference performance improved by 46%, open source solution breaks the length limit of LLM for multi-round conversations
@@ -361,7 +379,7 @@ Please visit our [documentation](https://www.colossalai.org/) and [examples](htt
...
@@ -361,7 +379,7 @@ Please visit our [documentation](https://www.colossalai.org/) and [examples](htt
## Installation
## Installation
Requirements:
Requirements:
- PyTorch >= 1.11 (PyTorch 2.x in progress)
- PyTorch >= 1.11 and PyTorch <= 2.1
- Python >= 3.7
- Python >= 3.7
- CUDA >= 11.0
- CUDA >= 11.0
-[NVIDIA GPU Compute Capability](https://developer.nvidia.com/cuda-gpus) >= 7.0 (V100/RTX20 and higher)
-[NVIDIA GPU Compute Capability](https://developer.nvidia.com/cuda-gpus) >= 7.0 (V100/RTX20 and higher)
...
@@ -495,11 +513,22 @@ This project is inspired by some related projects (some by our team and some by
...
@@ -495,11 +513,22 @@ This project is inspired by some related projects (some by our team and some by
To cite this project, you can use the following BibTeX citation.
To cite this project, you can use the following BibTeX citation.
```
```
@article{bian2021colossal,
@inproceedings{10.1145/3605573.3605613,
title={Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training},
author = {Li, Shenggui and Liu, Hongxin and Bian, Zhengda and Fang, Jiarui and Huang, Haichen and Liu, Yuliang and Wang, Boxiang and You, Yang},
author={Bian, Zhengda and Liu, Hongxin and Wang, Boxiang and Huang, Haichen and Li, Yongbin and Wang, Chuanrui and Cui, Fan and You, Yang},
title = {Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training},
journal={arXiv preprint arXiv:2110.14883},
year = {2023},
year={2021}
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},
parser.add_argument('--model_path',type=str,default="hpcai-tech/Colossal-LLaMA-2-7b-base",help="HF repo name or local path of the model")
parser.add_argument('--device',type=str,default="cuda:0",help="Set the device")
parser.add_argument('--max_new_tokens',type=int,default=512,help=" Set maximum numbers of tokens to generate, ignoring the number of tokens in the prompt")
parser.add_argument('--do_sample',type=bool,default=True,help="Set whether or not to use sampling")
parser.add_argument('--temperature',type=float,default=0.3,help="Set temperature value")
parser.add_argument('--top_k',type=int,default=50,help="Set top_k value for top-k-filtering")
parser.add_argument('--top_p',type=int,default=0.95,help="Set top_p value for generation")
parser.add_argument('--input_txt',type=str,default="明月松间照，",help="The prompt input to the model")
