dLLM: Simple Diffusion Language Modeling
This framework addresses the problem of scattered and non-transparent implementations of diffusion language models, making them more accessible and reproducible for researchers and developers.
This paper introduces dLLM, an open-source framework that unifies the core components of diffusion language modeling, including training, inference, and evaluation. It allows users to reproduce, finetune, deploy, and evaluate existing large diffusion language models and provides recipes for building small DLMs from scratch, converting BERT-style encoders or autoregressive LMs into DLMs.
Although diffusion language models (DLMs) are evolving quickly, many recent models converge on a set of shared components. These components, however, are distributed across ad-hoc research codebases or lack transparent implementations, making them difficult to reproduce or extend. As the field accelerates, there is a clear need for a unified framework that standardizes these common components while remaining flexible enough to support new methods and architectures. To address this gap, we introduce dLLM, an open-source framework that unifies the core components of diffusion language modeling -- training, inference, and evaluation -- and makes them easy to customize for new designs. With dLLM, users can reproduce, finetune, deploy, and evaluate open-source large DLMs such as LLaDA and Dream through a standardized pipeline. The framework also provides minimal, reproducible recipes for building small DLMs from scratch with accessible compute, including converting any BERT-style encoder or autoregressive LM into a DLM. We also release the checkpoints of these small DLMs to make DLMs more accessible and accelerate future research.