Is the Navier–Stokes Millennium problem solved?

No. Global smoothness or finite-time breakdown for the specified 3D problem remains unresolved.

What is already known?

Global finite-energy weak solutions exist, many conditional regularity criteria are known and possible singular sets are strongly restricted.

Why do simulations work if the theorem is open?

Numerical models can accurately handle many finite regimes without proving a universal statement for every allowed initial condition.

Would solving it make weather predictable?

Not perfectly. Regularity would not remove chaos, uncertain measurements or finite computing resources.

OpenMathematics

Can a smooth three-dimensional fluid flow become singular in finite time?

Navier–Stokes Existence and Smoothness

The immediate opportunity is not perfect weather prediction. It is faster, safer and more trustworthy simulation for engineering. A major proof could strengthen the theory behind those tools, but useful products can already be built today.

See what it could unlock

The question

Understand it in two minutes

The immediate opportunity is not perfect weather prediction. It is faster, safer and more trustworthy simulation for engineering. A major proof could strengthen the theory behind those tools, but useful products can already be built today.

A fluid carries its own motion with it. Fast-moving regions transport and twist the velocity field, while viscosity tries to smooth differences out. In two dimensions, mathematics can control this competition globally. In three dimensions, vortex stretching can transfer activity toward smaller scales, and present estimates do not rule out extreme concentration. The problem is not whether simulations can give useful answers. It is whether the equations are universally guaranteed to remain mathematically well behaved for every allowed initial condition.

01 · Step 1

The pattern

A fluid carries its own motion with it. Fast-moving regions transport and twist the velocity field, while viscosity tries to smooth differences out.

02 · Step 2

The translation

In two dimensions, mathematics can control this competition globally. In three dimensions, vortex stretching can transfer activity toward smaller scales, and present estimates do not rule out extreme concentration.

03 · Step 3

The open gap

The problem is not whether simulations can give useful answers. It is whether the equations are universally guaranteed to remain mathematically well behaved for every allowed initial condition.

Why people care

Foundations of fluid mathematics

The equations are central to continuum mechanics, yet their global three-dimensional regularity remains unknown.

New PDE techniques

Work on the problem advances harmonic analysis, geometric measure theory, numerical analysis and nonlinear PDE.

The structure behind turbulence

Regularity, energy transfer, intermittency and weak-solution behavior are deeply connected to questions raised by turbulent flow.

Possibilities

What it could unlock

What might become possible?

These are evidence-labeled possibilities, not predictions.

Scientific AIStrongly supported

Trustworthy Fluid Digital Twins

Simulation systems that expose confidence, numerical stability and out-of-distribution risk instead of returning an unexplained flow prediction.

Why it connects: Navier–Stokes regularity is a foundational theory question; useful assurance tooling does not need to wait for its solution.

See the possibility →

Scientific AIPlausible

Physics-Aware Surrogate Models

Fast learned approximations of expensive simulations, paired with conservation constraints and explicit uncertainty.

Why it connects: Navier–Stokes theory informs the domain, but practical surrogate models can be built and validated today.

See the possibility →

Ideas

Ideas it inspires

What could someone explore today?

Idea seeds for curiosity—not instructions or promises.

CFD, aerospace, energy and industrial simulation teamsStrongly supported

Simulation Trust Monitor

A monitoring layer that flags unstable residuals, mesh sensitivity, suspicious conservation error and out-of-distribution surrogate predictions.

Why it is thinkable now

Yes, as decision support—not as a universal mathematical certificate.

Aerospace, energy and industrial design teamsPlausible

Fluid Surrogate Platform

Train fast domain-specific approximations of validated CFD runs and return uncertainty and conservation diagnostics with every prediction.

Why it is thinkable now

Yes in controlled geometries and operating regimes.

Keep in mind

The honest boundary

Three-dimensional vortex stretching may outrun viscous smoothing

Energy estimates guarantee global weak solutions but do not control the scale-critical derivatives needed to exclude singularity or secure uniqueness for all smooth data.

What we know

The two-dimensional case is globally regular: The comparable incompressible problem in two spatial dimensions is globally well controlled; the unresolved difficulty is specifically three-dimensional. Global finite-energy weak solutions exist: Leray's framework provides global weak solutions from finite-energy initial data. Possible singularities are strongly restricted: The Caffarelli–Kohn–Nirenberg theorem shows that the singular set of suitable weak solutions is small in a precise parabolic-measure sense. Weak solutions can be nonunique in broad classes: Buckmaster and Vicol proved nonuniqueness for finite-energy weak solutions in a broad weak class, showing that admissibility and regularity matter.

Latest meaningful signal2019

Nonuniqueness is proved in a broad weak-solution class

Weak-solution behavior is shown to be more flexible than classical smooth-flow intuition suggests.

Sources behind this explanation 4

Nonuniqueness of weak solutions to the Navier–Stokes equationTristan Buckmaster, Vlad VicolSource verified Partial regularity of suitable weak solutions of the Navier–Stokes equationsLuis Caffarelli, Robert Kohn, Louis NirenbergSource verified Navier–Stokes EquationClay Mathematics InstituteSource verified Leray's fundamental work on the Navier–Stokes equations: a modern reviewWojciech S. Ożański, Benjamin C. PooleySource verified