Will Lean mathlib contain more than 10 million lines of code by 2030?
6
10kṀ20692030
59%
chance
1H
6H
1D
1W
1M
ALL
This question resolves based on the total number of lines of code (LOC) in the Lean mathlib repository.
As of late 2025, mathlib contains approximately 2.1 million lines of code across roughly 8,000 files. Over the past two years, growth has been approximately linear, which (if extrapolated) would not reach 10 million LOC by 2030.
Therefore, a “Yes” outcome implies a substantial acceleration in the rate of formalization, potentially driven by improved tooling, automation, or AI-assisted proof generation.
Resolution criterion:
This question resolves Yes if the official mathlib statistics report more than 10,000,000 total lines of code at any point on or before December 31, 2030. Otherwise, it resolves No.
Primary source:
This question is managed and resolved by Manifold.
Market context
Get
1,000 to start trading!
People are also trading
Related questions
Will an algorithm be able to work on million-line codebases before 2026?
17% chance
In 2029, will any AI be able to construct "reasonably" bug-free code of >= 10k LOC from a natural language specification? (Gary Marcus benchmark #4)
86% chance
Which theorems will be formally proven in Lean by the end of 2028?
Will the best public LLM at the end of 2025 solve more than 5 of the first 10 Project Euler problems published in 2026?
70% chance
What tactic will prove the most mathlib lemmas at the end of 2026?
Will Terence Tao write a paper with Lean code in it during the 2026 calendar year?
80% chance
Will reinforcement learning overtake LMs on math before 2028?
61% chance
Biggest formal math library in 2040 HoTT-based?
44% chance
Which MATH-AI 23 works will have >50 Google Scholar citations by end of 2026?
Will Kevin Buzzard successfully fulfill his grant specs of formalizing Fermat's Last Theorem in Lean within 5 years?
61% chance