English

Automatic Textbook Formalization

Artificial Intelligence 2026-04-06 v1

Abstract

We present a case study where an automatic AI system formalizes a textbook with more than 500 pages of graduate-level algebraic combinatorics to Lean. The resulting formalization represents a new milestone in textbook formalization scale and proficiency, moving from early results in undergraduate topology and restructuring of existing library content to a full standalone formalization of a graduate textbook. The formalization comprises 130K lines of code and 5900 Lean declarations and was conducted within one week by a total of 30K Claude 4.5 Opus agents collaborating in parallel on a shared code base via version control, simultaneously setting a record in multi-agent software engineering with usable results. The inference cost matches or undercuts what we estimate as the salaries required for a team of human experts, and we expect there is still the potential for large efficiencies to be made without the need for better models. We make our code, the resulting Lean code base and a side-by-side blueprint website available open-source.

Keywords

Cite

@article{arxiv.2604.03071,
  title  = {Automatic Textbook Formalization},
  author = {Fabian Gloeckle and Ahmad Rammal and Charles Arnal and Remi Munos and Vivien Cabannes and Gabriel Synnaeve and Amaury Hayat},
  journal= {arXiv preprint arXiv:2604.03071},
  year   = {2026}
}

Comments

19 pages

R2 v1 2026-07-01T11:52:54.367Z