Making large AI models cheaper, faster, and more accessible.

A simple, performant and scalable Jax LLM!

Mesh TensorFlow: Model Parallelism Made Easier.

veRL is a flexible and efficient RL framework for LLMs.

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

A Native-PyTorch Library for LLM Fine-tuning.