• [2022/03/21] Supporting efficient large model training on AMD Instinct GPUs with DeepSpeed
• [2022/03/07] Maximizing Communication Efficiency for Large-scale Training via 0/1 Adam
• [2022/01/19] DeepSpeed: Advancing MoE inference and training to power next-generation AI scale
  • Mixture of Experts (MoE) for NLG tutorial.
  • Mixture of Experts (MoE) Inference tutorial.
• [2021/11/15] Autotuning: Automatically discover the optimal DeepSpeed configuration that delivers good training speed
• [2021/10/11] Using DeepSpeed and Megatron to Train Megatron-Turing NLG 530B, the World’s Largest and Most Powerful Generative Language Model
  • Read more on how to train large models with DeepSpeed

DeepSpeed is a deep learning optimization library that makes distributed training easy, efficient, and effective.

10x Larger Models

10x Faster Training

Minimal Code Change

DeepSpeed delivers extreme-scale model training for everyone, from data scientists training on massive supercomputers to those training on low-end clusters or even on a single GPU:

• Extreme scale: Using current generation of GPU clusters with hundreds of devices, 3D parallelism of DeepSpeed can efficiently train deep learning models with trillions of parameters.
• Extremely memory efficient: With just a single GPU, ZeRO-Offload of DeepSpeed can train models with over 10B parameters, 10x bigger than the state of arts, democratizing multi-billion-parameter model training such that many deep learning scientists can explore bigger and better models.
• Extremely long sequence length: Sparse attention of DeepSpeed powers an order-of-magnitude longer input sequence and obtains up to 6x faster execution comparing with dense transformers.
• Extremely communication efficient: 3D parallelism improves communication efficiency allows users to train multi-billion-parameter models 2–7x faster on clusters with limited network bandwidth. 1-bit Adam, 0/1 Adam and 1-bit LAMB reduce communication volume by up to 26x while achieving similar convergence efficiency to Adam/LAMB, allowing for scaling to different types of GPU clusters and networks.

Early adopters of DeepSpeed have already produced a language model (LM) with over 17B parameters called Turing-NLG, establishing a new SOTA in the LM category.

DeepSpeed is an important part of Microsoft’s new AI at Scale initiative to enable next-generation AI capabilities at scale, where you can find more information here.

For further documentation, tutorials, and technical deep-dives please see deepspeed.ai!

Training advanced deep learning models is challenging. Beyond model design, model scientists also need to set up the state-of-the-art training techniques such as distributed training, mixed precision, gradient accumulation, and checkpointing. Yet still, scientists may not achieve the desired system performance and convergence rate. Large model sizes are even more challenging: a large model easily runs out of memory with pure data parallelism and it is difficult to use model parallelism. DeepSpeed addresses these challenges to accelerate model development and training.

The quickest way to get started with DeepSpeed is via pip, this will install the latest release of DeepSpeed which is not tied to specific PyTorch or CUDA versions. DeepSpeed includes several C /CUDA extensions that we commonly refer to as our 'ops'. By default, all of these extensions/ops will be built just-in-time (JIT) using torch's JIT C extension loader that relies on ninja to build and dynamically link them at runtime.

Note: PyTorch must be installed before installing DeepSpeed.

pip install deepspeed

After installation, you can validate your install and see which extensions/ops your machine is compatible with via the DeepSpeed environment report.

ds_report

If you would like to pre-install any of the DeepSpeed extensions/ops (instead of JIT compiling) or install pre-compiled ops via PyPI please see our advanced installation instructions.

On Windows you can build wheel with following steps, currently only inference mode is supported.

1. Install pytorch, such as pytorch 1.8 cuda 11.1
2. Install visual cpp build tools, such as VS2019 C x64/x86 build tools
3. Launch cmd console with Administrator privilege for creating required symlink folders
4. Run python setup.py bdist_wheel to build wheel in dist folder

Below we provide a brief feature list, see our detailed feature overview for descriptions and usage.

• Distributed Training with Mixed Precision
  • 16-bit mixed precision
  • Single-GPU/Multi-GPU/Multi-Node
• Model Parallelism
  • Support for Custom Model Parallelism
  • Integration with Megatron-LM
• Pipeline Parallelism
  • 3D Parallelism
• The Zero Redundancy Optimizer (ZeRO)
  • Optimizer State and Gradient Partitioning
  • Activation Partitioning
  • Constant Buffer Optimization
  • Contiguous Memory Optimization
• ZeRO-Offload
  • Leverage both CPU/GPU memory for model training
  • Support 10B model training on a single GPU
• Ultra-fast dense transformer kernels
• Sparse attention
  • Memory- and compute-efficient sparse kernels
  • Support 10x longer sequences than dense
  • Flexible support to different sparse structures
• 1-bit Adam, 0/1 Adam and 1-bit LAMB
  • Custom communication collective
  • Up to 26x communication volume saving
• Additional Memory and Bandwidth Optimizations
  • Smart Gradient Accumulation
  • Communication/Computation Overlap
• Training Features
  • Simplified training API
  • Gradient Clipping
  • Automatic loss scaling with mixed precision
• Training Optimizers
  • Fused Adam optimizer and arbitrary torch.optim.Optimizer
  • Memory bandwidth optimized FP16 Optimizer
  • Large Batch Training with LAMB Optimizer
  • Memory efficient Training with ZeRO Optimizer
  • CPU-Adam
• Training Agnostic Checkpointing
• Advanced Parameter Search
  • Learning Rate Range Test
  • 1Cycle Learning Rate Schedule
• Simplified Data Loader
• Curriculum Learning
  • A curriculum learning-based data pipeline that presents easier or simpler examples earlier during training
  • Stable and 3.3x faster GPT-2 pre-training with 8x/4x larger batch size/learning rate while maintaining token-wise convergence speed
  • Complementary to many other DeepSpeed features
• Performance Analysis and Debugging
• Mixture of Experts (MoE)

All DeepSpeed documentation can be found on our website: deepspeed.ai

DeepSpeed welcomes your contributions! Please see our contributing guide for more details on formatting, testing, etc.

This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit https://cla.opensource.microsoft.com.

When you submit a pull request, a CLA bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., status check, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.

This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact [email protected] with any additional questions or comments.

1. Samyam Rajbhandari, Jeff Rasley, Olatunji Ruwase, Yuxiong He. (2019) ZeRO: memory optimizations toward training trillion parameter models. arXiv:1910.02054 and In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis (SC '20).
2. Jeff Rasley, Samyam Rajbhandari, Olatunji Ruwase, and Yuxiong He. (2020) DeepSpeed: System Optimizations Enable Training Deep Learning Models with Over 100 Billion Parameters. In Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD '20, Tutorial).
3. Minjia Zhang, Yuxiong He. (2020) Accelerating Training of Transformer-Based Language Models with Progressive Layer Dropping. arXiv:2010.13369 and NeurIPS 2020.
4. Jie Ren, Samyam Rajbhandari, Reza Yazdani Aminabadi, Olatunji Ruwase, Shuangyan Yang, Minjia Zhang, Dong Li, Yuxiong He. (2021) ZeRO-Offload: Democratizing Billion-Scale Model Training. arXiv:2101.06840.
5. Hanlin Tang, Shaoduo Gan, Ammar Ahmad Awan, Samyam Rajbhandari, Conglong Li, Xiangru Lian, Ji Liu, Ce Zhang, Yuxiong He. (2021) 1-bit Adam: Communication Efficient Large-Scale Training with Adam's Convergence Speed. arXiv:2102.02888 and ICML 2021.
6. Samyam Rajbhandari, Olatunji Ruwase, Jeff Rasley, Shaden Smith, Yuxiong He. (2021) ZeRO-Infinity: Breaking the GPU Memory Wall for Extreme Scale Deep Learning. arXiv:2104.07857.
7. Conglong Li, Ammar Ahmad Awan, Hanlin Tang, Samyam Rajbhandari, Yuxiong He. (2021) 1-bit LAMB: Communication Efficient Large-Scale Large-Batch Training with LAMB's Convergence Speed. arXiv:2104.06069.
8. Conglong Li, Minjia Zhang, Yuxiong He. (2021) Curriculum Learning: A Regularization Method for Efficient and Stable Billion-Scale GPT Model Pre-Training. arXiv:2108.06084.
9. Yucheng Lu, Conglong Li, Minjia Zhang, Christopher De Sa, Yuxiong He. (2022) Maximizing Communication Efficiency for Large-scale Training via 0/1 Adam. arXiv:2202.06009.
10. Samyam Rajbhandari, Conglong Li, Zhewei Yao, Minjia Zhang, Reza Yazdani Aminabadi, Ammar Ahmad Awan, Jeff Rasley, Yuxiong He. (2022) DeepSpeed-MoE: Advancing Mixture-of-Experts Inference and Training to Power Next-Generation AI Scale arXiv:2201.05596.

1. DeepSpeed KDD 2020 Tutorial
   1. Overview
   2. ZeRO large model training
   3. 17B T-NLG demo
   4. Fastest BERT training RScan tuning
   5. DeepSpeed hands on deep dive: part 1, part 2, part 3
   6. FAQ
2. Microsoft Research Webinar
   • Registration is free and all videos are available on-demand.
   • ZeRO & Fastest BERT: Increasing the scale and speed of deep learning training in DeepSpeed.
3. DeepSpeed on AzureML
4. Community Tutorials
   • DeepSpeed: All the tricks to scale to gigantic models
   • Turing-NLG, DeepSpeed and the ZeRO optimizer