RetentionLabs

dLLM: Simple Diffusion Language Modeling

🚀 Efficient implementations of state-of-the-art linear attention models

A Reproduction of GDM's Nested Learning Paper

🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level Computation (NeurIPS 2025)