1 repo
Awesome GitHub RepositoriesModel Parallelism Frameworks
Techniques for splitting neural network layers across multiple devices to manage memory and compute load.
Distinguishing note: Focuses on pipeline-based layer partitioning, distinct from data-parallel training approaches.
Explore 1 awesome GitHub repository matching artificial intelligence & ml · Model Parallelism Frameworks. Refine with filters or upvote what's useful.
Awesome Model Parallelism Frameworks GitHub Repositories
Find the best repos with AI.We'll search the best matching repositories with AI.
- deepspeedai/DeepSpeed
deepspeedai/DeepSpeed
41,638View on GitHubDeepSpeed is a high-performance library designed to scale deep learning model training and inference across massive clusters of GPUs and compute nodes. It provides a comprehensive suite of tools for distributed training, enabling the execution of models that exceed the memory capacity of single devices through advanced parameter partitioning, pipeline-based model parallelism, and memory-efficient state offloading. The framework distinguishes itself through specialized communication-efficient optimizers and hardware-aware acceleration techniques. By utilizing gradient compression, quantization
Neural network layers are partitioned into sequential stages across multiple devices to distribute memory load and enable large-scale model training.
Pythonbillion-parameterscompressiondata-parallelism
