This is a dynamic list and may never be able to satisfy particular standards for completeness. You can help by editing the page to add missing items, with references to reliable sources.
This list is a composite of notable unsolved problems mentioned in previously published lists, including but not limited to lists considered authoritative, and the problems listed here vary widely in both difficulty and importance.
Notable lists
For over a century, various mathematicians and organizations have published and promoted lists of unsolved mathematical problems. In some cases, the lists have been associated with prizes for the discoverers of solutions, with the Millennium Prize Problems each containing a reward of one million dollars.
Contrary to most of the Hilbert problems and many other problems described in this article which allow for varying interpretations, the Millennium Prize problems are precisely described to provide clear conditions to award winners. After Perelman's resolution of the Poincaré conjecture, the official descriptions of the remaining six problems were codified by six specialists and released publicly. While the problems can touch upon greater research topics, and can themselves be only special cases of more general open problems, the Clay Institute decided upon these particular cases of the problems:
The Poincaré conjecture can be formulated in terms of the generalized Poincaré conjecture for different categories of manifolds. In the category of piecewise linear manifolds, the generalized conjecture is still an open problem for 4-dimensional manifolds. For differential manifolds, it is known to be false generally, but it is an open problem to determine in which dimensions it is true. In other words, it is not yet fully known on which spheres there exist no exotic smooth structures. The Millennium Prize was offered for the topological version of the conjecture, that remained open only for 3-dimensional manifolds during the publication of the list.
The Birch and Swinnerton-Dyer conjecture can be formulated for elliptic curves over any number field, and generalizations for more general arithmetic objects have been proposed. The Millennium Prize is offered for resolving the conjecture for elliptic curves over rational numbers.
The Navier–Stokes equations have a general form that can take into account many physical phenomena including the difference between volume viscosity and dynamic viscosity, compressible flow, change of viscosities in the flow, and an external force. The Millennium Prize is offered for proving any one of four statements, two of which deal with the Navier–Stokes equations with no external force while the other two deal with the Navier–Stokes equations with a smooth external force.
The Riemann hypothesis has many generalizations for different types of L-functions that are expected to be true. The Millennium Prize problem is offered for showing the original Riemann hypothesis for the Riemann zeta function.
Notebooks
The Kourovka Notebook (Russian: Коуровская тетрадь) is a collection of unsolved problems in group theory, first published in 1965 and updated many times since.
The Sverdlovsk Notebook (Russian: Свердловская тетрадь) is a collection of unsolved problems in semigroup theory, first published in 1965 and updated every 2 to 4 years since.
Casas-Alvero conjecture: if a polynomial of degree defined over a field of characteristic has a factor in common with its first through -th derivative, then must be the -th power of a linear polynomial?
Williamson conjecture: the problem of finding Williamson matrices, which can be used to construct Hadamard matrices.
Lander's conjecture: if an abelian group of order contains a difference set of order , and a prime divides both and , must the Sylow -subgroup be non-cyclic? This implies Ryser's cyclic difference-set conjecture, and hence the circulant Hadamard conjecture.
Pierce–Birkhoff conjecture: every piecewise-polynomial is the maximum of a finite set of minimums of finite collections of polynomials.
Rota's basis conjecture: for matroids of rank with disjoint bases , it is possible to create an matrix whose rows are and whose columns are also bases.
The free Burnside group is finite; in its Cayley graph, shown here, each of its 27 elements is represented by a vertex. The question of which other groups are finite remains open.
Bounded Burnside problem: for which positive integers m, n is the free Burnside group B(m,n) finite? In particular, is B(2, 5) finite?
Guralnick–Thompson conjecture on the composition factors of groups in genus-0 systems
Herzog–Schönheim conjecture: if a finite system of left cosets of subgroups of a group form a partition of , then the finite indices of said subgroups cannot be distinct.
The inverse Galois problem: is every finite group the Galois group of a Galois extension of the rationals?
The Brennan conjecture: estimating the integral powers of the moduli of the derivative of conformal maps into the open unit disk, on certain subsets of
The Dittert conjecture concerning the maximum achieved by a particular function of matrices with real, nonnegative entries satisfying a summation condition
The lonely runner conjecture – if runners with pairwise distinct speeds run round a track of unit length, will every runner be "lonely" (that is, be at least a distance from each other runner) at some time?
Superpermutations - smallest possible string of n digits that contains all possible permutations of n
Map folding – various problems in map folding and stamp folding.
No-three-in-line problem – how many points can be placed in the grid so that no three of them lie on a line?
The sunflower conjecture – can the number of size sets required for the existence of a sunflower of sets be bounded by an exponential function in for every fixed ?
Frankl's union-closed sets conjecture – for any family of sets closed under sums there exists an element (of the underlying space) belonging to half or more of the sets
Does every positive integer generate a juggler sequence terminating at 1?
Lyapunov function: Lyapunov's second method for stability – For what classes of ODEs, describing dynamical systems, does Lyapunov's second method, formulated in the classical and canonically generalized forms, define the necessary and sufficient conditions for the (asymptotical) stability of motion?
Given the width of a tic-tac-toe board, what is the smallest dimension such that X is guaranteed to have a winning strategy? (See also Hales–Jewett theorem and nd game)
Borsuk's problem on upper and lower bounds for the number of smaller-diameter subsets needed to cover a boundedn-dimensional set.
The covering problem of Rado: if the union of finitely many axis-parallel squares has unit area, how small can the largest area covered by a disjoint subset of squares be?
The Erdős–Oler conjecture: when is a triangular number, packing circles in an equilateral triangle requires a triangle of the same size as packing circles.
Reinhardt's conjecture: the smoothed octagon has the lowest maximum packing density of all centrally-symmetric convex plane sets
Sphere packing problems, including the density of the densest packing in dimensions other than 1, 2, 3, 8 and 24, and its asymptotic behavior for high dimensions.
Closed curve problem: find (explicit) necessary and sufficient conditions that determine when, given two periodic functions with the same period, the integral curve is closed.
The filling area conjecture, that a hemisphere has the minimum area among shortcut-free surfaces in Euclidean space whose boundary forms a closed curve of given length
The Hopf conjectures relating the curvature and Euler characteristic of higher-dimensional Riemannian manifolds
In three dimensions, the kissing number is 12, because 12 non-overlapping unit spheres can be put into contact with a central unit sphere. (Here, the centers of outer spheres form the vertices of a regular icosahedron.) Kissing numbers are only known exactly in dimensions 1, 2, 3, 4, 8 and 24.
The big-line-big-clique conjecture on the existence of either many collinear points or many mutually visible points in large planar point sets
The Dirac–Motzkin conjecture on the minimum number of ordinary lines for n points in the Euclidean plane
The Hadwiger conjecture on covering n-dimensional convex bodies with at most 2n smaller copies
Bellman's lost-in-a-forest problem – find the shortest route that is guaranteed to reach the boundary of a given shape, starting at an unknown point of the shape with unknown orientation
Borromean rings — are there three unknotted space curves, not all three circles, which cannot be arranged to form this link?
The Kelvin problem on minimum-surface-area partitions of space into equal-volume cells, and the optimality of the Weaire–Phelan structure as a solution to the Kelvin problem
Can every spherical non-convex polyhedron that tiles space by translation have its faces grouped into patches with the same combinatorial structure as a parallelohedron?
Does every higher-dimensional tiling by translations of convex polytope tiles have an affine transformation taking it to a Voronoi diagram?
Hilbert's third problem for non-Euclidean geometries: in spherical or hyperbolic geometry, must polyhedra with the same volume and Dehn invariant be scissors-congruent?
Suppose Alice has a winning strategy for the vertex coloring game on a graph with colors. Does she have one for colors?
Graph coloring and labeling
An instance of the Erdős–Faber–Lovász conjecture: a graph formed from four cliques of four vertices each, any two of which intersect in a single vertex, can be four-colored.
Turán's brick factory problem – Is there a drawing of any complete bipartite graph with fewer crossings than the number given by Zarankiewicz?
Guy's conjecture on the crossing number for complete graphs – Is there a drawing of any complete graph with fewer crossings than the number given by his upper bound?
The Oberwolfach problem on which 2-regular graphs have the property that a complete graph on the same number of vertices can be decomposed into edge-disjoint copies of the given graph.
Characterise word-representable near-triangulations containing the complete graph K4 (such a characterisation is known for K4-free planar graphs)
Classify graphs with representation number 3, that is, graphs that can be represented using 3 copies of each letter, but cannot be represented using 2 copies of each letter
Which (hard) problems on graphs can be translated to words representing them and solved on words (efficiently)?
Miscellaneous graph theory
Delta-conjecture (1978): consider a complete graph each edge of which is colored by one of colors such that there exists a triangle colored in three pairwise distinct colors. Then, in each chromatic component , one can choose a maximal independent vertex-set such that the intersection of the obtained sets is empty.
The imbalance conjecture: If the imbalance for each edge of a graph is at least 1, is the multiset of all edge imbalances always graphic?
The second neighborhood problem: does every oriented graph contain a vertex for which there are at least as many other vertices at distance two as at distance one?
The main gap conjecture, e.g. for uncountable first order theories, for AECs, and for -saturated models of a countable theory.
Shelah's categoricity conjecture for : If a sentence is categorical above the Hanf number then it is categorical in all cardinals above the Hanf number.
Shelah's eventual categoricity conjecture: For every cardinal there exists a cardinal such that if an AECK with LS(K) is categorical in a cardinal above then it is categorical in all cardinals above .
The stable field conjecture: every infinite field with a stable first-order theory is separably closed.
The universality problem for C-free graphs: For which finite sets C of graphs does the class of C-free countable graphs have a universal member under strong embeddings?
The universality spectrum problem: Is there a first-order theory whose universality spectrum is minimum?
Assume K is the class of models of a countable first order theory omitting countably many types. If K has a model of cardinality does it have a model of cardinality continuum?
Does a finitely presented homogeneous structure for a finite relational language have finitely many reducts?
Does there exist an o-minimal first order theory with a trans-exponential (rapid growth) function?
If the class of atomic models of a complete first order theory is categorical in the , is it categorical in every cardinal?
Is every infinite, minimal field of characteristic zero algebraically closed? (Here, "minimal" means that every definable subset of the structure is finite or co-finite.)
Is the Borel monadic theory of the real order (BMTO) decidable? Is the monadic theory of well-ordering (MTWO) consistently decidable?
Is the theory of the field of Laurent series over decidable? of the field of polynomials over ?
Is there a logic L which satisfies both the Beth property and Δ-interpolation, is compact but does not satisfy the interpolation property?
6 is a perfect number because it is the sum of its proper positive divisors, 1, 2 and 3. It is not known how many perfect numbers there are, nor if any of them is odd.
Büchi's problem on sufficiently large sequences of square numbers with constant second difference.
Minimum overlap problem of estimating the minimum possible maximum number of times a number appears in the termwise difference of two equally large sets partitioning the set
Manin conjecture: if K-rational points on Fano variety are Zariski-dense subset, then the distribution of points of height: in any Zariski-open subset is proportional to , where is rank of Picard group of that variety.
Sato–Tate conjecture: also a number of related conjectures that are generalizations of the original conjecture.
n conjecture: a generalization of the abc conjecture to more than three integers.
abc conjecture: for any , is true for only finitely many positive such that .
Szpiro's conjecture: for any , there is some constant such that, for any elliptic curve defined over with minimal discriminant and conductor , we have .
How well can non-quadratic irrational numbers be approximated? What is the irrationality measure of specific (suspected) transcendental numbers such as and ?
Goldbach's conjecture states that all even integers greater than 2 can be written as the sum of two primes. Here this is illustrated for the even integers from 4 to 28.
Bunyakovsky conjecture: if an integer-coefficient polynomial has a positive leading coefficient, is irreducible over the integers, and has no common factors over all where is a positive integer, then is prime infinitely often.
Dickson's conjecture: for a finite set of linear forms with each , there are infinitely many for which all forms are prime, unless there is some congruence condition preventing it.
Dubner's conjecture: every even number greater than is the sum of two primes which both have a twin.
The Gaussian moat problem: is it possible to find an infinite sequence of distinct Gaussian prime numbers such that the difference between consecutive numbers in the sequence is bounded?
New Mersenne conjecture: for any odd natural number , if any two of the three conditions or , is prime, and is prime are true, then the third condition is also true.
Schinzel's hypothesis H that for every finite collection of nonconstant irreducible polynomials over the integers with positive leading coefficients, either there are infinitely many positive integers for which are all primes, or there is some fixed divisor which, for all , divides some .
For any given integer a > 0, are there infinitely many Lucas–Wieferich primes associated with the pair (a, −1)? (Specially, when a = 1, this is the Fibonacci-Wieferich primes, and when a = 2, this is the Pell-Wieferich primes)
For any given integer a > 0, are there infinitely many primes p such that ap − 1 ≡ 1 (mod p2)?
For any given integer b which is not a perfect power and not of the form −4k4 for integer k, are there infinitely many repunit primes to base b?
For any given integers , with gcd(k, c) = 1 and gcd(b, c) = 1, are there infinitely many primes of the form with integer n ≥ 1?
Disproof of the Connes embedding problem (Zhengfeng Ji, Anand Natarajan, Thomas Vidick, John Wright, and Henry Yuen, 2020), as a consequence of their proof that .
Group theory
Existence of a non-sofic, discrete, countable group (solution in an anonymized paper published by OpenAI, 2026).
Disproof of Jaeger's Petersen-coloring conjecture: every bridgeless cubic graph has a cycle-continuous mapping to the Petersen graph (NeuralReformist (July 2026); Jan Goedgebeur, Jorik Jooken, Edita Máčajová, Davide Mattiolo, Giuseppe Mazzuoccolo, and Nikolay Ulyanov, August 2026).