veRL is a flexible and efficient RL framework for LLMs.
DeepSpeed version of NVIDIA's Megatron-LM that adds additional support for several features such as MoE model training, Curriculum Learning, 3D Parallelism, and others.
Mesh TensorFlow: Model Parallelism Made Easier.
A simple, performant and scalable Jax LLM!
A native PyTorch Library for large model training.
A Native-PyTorch Library for LLM Fine-tuning.
An implementation of model parallel autoregressive transformers on GPUs, based on the DeepSpeed library.
Your email address will not be published. Required fields are marked *
Comment *
Name *
Email *
Website
Captcha: 13 + 15 = ?*
Save my name, email, and website in this browser for the next time I comment.
DeepSpeed version of NVIDIA's Megatron-LM that adds additional support for several features such as MoE model training, Curriculum Learning, 3D Parallelism, and others.