English

Modalities, a PyTorch-native Framework For Large-scale LLM Training and Research

Machine Learning 2026-02-10 v1 Distributed, Parallel, and Cluster Computing

Abstract

Today's LLM (pre-) training and research workflows typically allocate a significant amount of compute to large-scale ablation studies. Despite the substantial compute costs of these ablations, existing open-source frameworks provide limited tooling for these experiments, often forcing researchers to write their own wrappers and scripts. We propose Modalities, an end-to-end PyTorch-native framework that integrates data-driven LLM research with large-scale model training from two angles. Firstly, by integrating state-of-the-art parallelization strategies, it enables both efficient pretraining and systematic ablations at trillion-token and billion-parameter scale. Secondly, Modalities adopts modular design with declarative, self-contained configuration, enabling reproducibility and extensibility levels that are difficult to achieve out-of-the-box with existing LLM training frameworks.

Keywords

Cite

@article{arxiv.2602.08387,
  title  = {Modalities, a PyTorch-native Framework For Large-scale LLM Training and Research},
  author = {Max Lübbering and Timm Ruland and Richard Rutmann and Felix Stollenwerk and David Fitzek and Michael Fromm and Alexander Weber and Rafet Sifa and Nicolas Flores-Herr and Joachim Köhler and Mehdi Ali},
  journal= {arXiv preprint arXiv:2602.08387},
  year   = {2026}
}
R2 v1 2026-07-01T10:27:28.912Z