DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.
Repository Index
89,572 visible repositories in the index
1–12 of 12 · Most stars
TranslatedPArallel Distributed Deep LEarning: Machine Learning Framework from Industrial Practice (Core framework of 'PaddlePaddle' (飞桨, Feijiang), for deep learning & machine learning: high-performance single-machine and distributed training, and cross-platform deployment)
- paddlepaddle
- deep-learning
- scalability
- machine-learning
Python package built to ease deep learning on graph, on top of existing DL frameworks.
- deep-learning
- graph-neural-networks
A Neural Net Training Interface on TensorFlow, with focus on speed + flexibility
A Next-Generation Training Engine Built for Ultra-Large MoE Models
A unified inference and post-training framework for accelerated video generation.
- diffusers
- diffusion-models
- video-generation
- distillation
MindSpore is a new open source deep learning training/inference framework that could be used for mobile, edge and cloud scenarios.
FEDML - The unified and scalable ML library for large-scale distributed training, model serving, and federated learning. FEDML Launch, a cross-cloud scheduler, further enables running any AI jobs on any GPU cloud or on-premise cluster. Built on this library, TensorOpera AI (https://TensorOpera.ai) is your generative AI platform at scale.
- federated-learning
- deep-learning
- distributed-training
- edge-ai
Open-source deep-learning framework for building, training, and fine-tuning deep learning models using state-of-the-art Physics-ML methods
- deep-learning
- machine-learning
- nvidia-gpu
- physics
Miles is an enterprise-facing reinforcement learning framework for LLM and VLM post-training, forked from and co-evolving with slime.
Decentralized deep learning in PyTorch. Built to train models on thousands of volunteers across the world.
- deep-learning
- pytorch
- volunteer-computing
- mixture-of-experts
Whole-body motion tracking for the Unitree G1 in under a minute. BlinkMimic's world-model trainer rebuilt in JAX, 3.7x faster on four H100s.