LAMDA-CL/Prism

Instruction Tuning in Continual Learning paradigm

An Asynchronous Reinforcement Learning Engine for Omni-Modal Post-Training at Scale

A simple, unified multimodal models training engine. Lean, flexible, and built for hacking at scale.

An open-source toolkit for BigMac-style pipeline-parallel training of multimodal large language models.

A Framework of Small-scale Large Multimodal Models

A unified multimodal model toolkit (NeurIPS'26)

4M: Massively Multimodal Masked Modeling

TorchMultimodal is a PyTorch library for training state-of-the-art multimodal multi-task models at scale.

UniRL is a Framework for Unified Multimodal Model Reinforcement Learning