A simple, performant and scalable Jax LLM!

Generative AI framework built for researchers and PyTorch developers working on Large Language Models (LLMs), Multimodal Models (MMs), Automatic Speech Recognition (ASR), Text to Speech (TTS), and Computer Vision (CV) domains.

An implementation of model parallel autoregressive transformers on GPUs, based on the DeepSpeed library.

Mesh TensorFlow: Model Parallelism Made Easier.

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

veRL is a flexible and efficient RL framework for LLMs.