Contributor workspace
Sign in to claim tasks, verify results and earn reputation. Browsing is open to everyone.
Sign inFor every viscosity the 166-page manuscript constructs a force and a smooth solution on with , bounded kinetic energy and , which is Fefferman's alternative (C) on ; a corollary gives alternative (D) on . A Lean 4 project formalizes the (C) and (D) statements and an unforced Euler blow-up. If correct, it settles the Clay problem as stated through the forced alternatives; it says nothing about the unforced equations ().
Unrefereed. An independent rebuild of the Lean project (Zenodo, 10 Oct 2026) found only the standard axioms and statements matching DeepMind's Formal Conjectures, but Bastounis–Circelli–Hansen (arXiv:2610.08144) show the Lean code proves different, sometimes weaker, intermediate bounds than the prose and so does not certify the written proof; they make no claim about its correctness. The Clay Institute (11 Sep 2026) called the problem 'apparently settled' and announced a deliberately unhurried evaluation; no prize has been awarded.
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.
Independent rebuild reports only propext, Classical.choice and Quot.sound, with no sorry or native_decide in 816 files.
The same audit reports that all 16 definitions and both main statements agree with DeepMind's Formal Conjectures encoding (commit 8bf45ed).
The Analysis & PDE 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.
A PDE specialist should compare the Lean definitions (force class, smoothness, energy bound, torus normalization) line by line with the official Clay problem description. The agreement check above only compares two formalizations with each other.
Bastounis–Circelli–Hansen find the Lean code proves different intermediate estimates by different arguments (e.g. an inverse estimate with m+5 rather than m+4 derivatives, and a pressure-flux bound that also depends on A_R). The formal theorem stands, but the prose argument is not certified by it.
No independent group has published a full check of the written argument. Work can be split by section: the self-similar ansatz, stability of the collapsing core, and the forcing construction.
Clay's rules require publication in a refereed venue and two years of general acceptance. As of 16 Oct 2026 the manuscript has not been submitted or accepted anywhere publicly.