Contributor workspace
Sign in to claim tasks, verify results and earn reputation. Browsing is open to everyone.
Sign inAbout 13 million lines of Lean following the Darmon–Diamond–Taylor exposition of Wiles–Taylor–Wiles prove Mathlib's statement of FLT, with small exponents handled via existing regular-prime work. It compiles about twenty times slower than Mathlib and is not integrated with it. It certifies a known theorem; it does not provide the reusable, Mathlib-integrated Galois-representation infrastructure that the Imperial project targets.
Kevin Buzzard compiled the repository, ran Comparator against the FLT statement ('it checks out') and inspected non-theorem code flagged by the agents (4 Sep 2026).
Verification broken into steps anyone can take. Pick an open one, do it, and submit what you found below. Matching how a check is settled earns +8; catching a failure earns +20.
Buzzard compiled the repository and Comparator accepted the statement against Mathlib's FermatLastTheorem.
Buzzard checked the code the agents flagged as not being theorems (a convenience tactic) and found nothing that changes the statement.
The Number Theory committee has no members yet, so this claim can be machine-checked and verified by anyone but not committee-reviewed.
No verifications yet.
Sign in to verify.
Buzzard's post says the repository proves the whole argument, including Fontaine theory and Mazur's Eisenstein-ideal results, but does not say how Langlands–Tunnell and Ribet's level lowering are handled. A dependency map would settle this.
Publish a clean-room build log with #print axioms FermatLastTheorem output from a party other than the author and Buzzard.