← All problems

Twin prime conjecture

180 years old · traced to

A twin prime pair is a pair of primes (p,p+2)(p,p+2). Let p1<p2<⋯p_1<p_2<\cdots be the primes in increasing order.

Conjecture. There are infinitely many primes pp such that p+2p+2 is also prime:

#{p prime:p+2 prime}=∞.\#\{p \text{ prime} : p+2 \text{ prime}\}=\infty .

Every prime after 22 is odd, so pn+1−pn≥2p_{n+1}-p_n\ge2 for n≥2n\ge2. The conjecture is therefore equivalent to

lim inf⁡n→∞ (pn+1−pn)=2.\liminf_{n\to\infty}\,(p_{n+1}-p_n)=2 .

A solution must prove that infinitely many twin prime pairs exist. A bound on the lim inf⁡\liminf larger than 22 does not suffice. [12]

Hardy and Littlewood conjectured the stronger asymptotic

π2(x)=#{p≤x:p+2 prime}∼2C2∫2xdt(log⁡t)2,C2=∏p>2(1−1(p−1)2)≈0.6602,\pi_2(x)=\#\{p\le x: p+2 \text{ prime}\}\sim 2C_2\int_2^x\frac{dt}{(\log t)^2},\qquad C_2=\prod_{p>2}\Bigl(1-\frac{1}{(p-1)^2}\Bigr)\approx0.6602 ,

which is not required for the conjecture [1].

Background

The twin prime problem is the simplest case of the general question of when several linear forms are simultaneously prime. That question is formalized by Dickson's conjecture and by the Hardy–Littlewood prime kk-tuples conjecture [1]. It is also the case k=2k=2 of de Polignac's 1849 conjecture that every even number occurs infinitely often as a gap between consecutive primes.

Brun introduced his sieve to attack the problem in 1915–1919. He showed that the sum of the reciprocals of the twin primes converges, so twin primes are sparse compared with all primes [11]. Sieve methods in general face the parity barrier. They cannot by themselves distinguish primes from products of two primes, and Polymath showed that the bound 66 under the generalized Elliott–Halberstam conjecture is the limit of purely sieve-theoretic arguments of this type [6]. Modern progress instead studies the gap bound

H1=lim inf⁡n→∞ (pn+1−pn),H_1=\liminf_{n\to\infty}\,(p_{n+1}-p_n),

which the conjecture asserts equals 22.

Equivalent formulations
  1. A twin prime is a pair of primes differing by 22. Twin prime conjecture. There are infinitely many primes qq such that q+2q+2 is prime. The conjecture is one of Landau's four problems; the source claims it is proved, but the supplied parser status gives no resolution evidence.

    source: Agostino Prástaro, “The Landau's problems. I-II”, arXiv:1208.2473 (2015).

  2. Let pnp_n denote the nnth prime and let dn=pn+1−pnd_n=p_{n+1}-p_n. The twin prime conjecture.

    lim inf⁡n→∞dn=2.\liminf\limits_{n \to \infty} d_n=2.

    This is the assertion that the gap 22 occurs infinitely often between consecutive primes. It remains open.

    source: Janos Pintz, “Polignac Numbers, Conjectures of Erdös on Gaps between Primes, Arithmetic Progressions in Primes, and the Bounded Gap Conjecture”, arXiv:1305.6289 (2013).

  3. A twin prime is a prime number that differs from another prime number by two. Twin prime conjecture. There are infinitely many pairs of twin primes. The conjecture is a famous open problem in number theory; although bounded gaps between infinitely many primes are known, no proof that infinitely many prime pairs differ by exactly two is currently known.

    source: Vineet Kumar, “Prime number generation and factor elimination”, arXiv:1411.3356 (2014).

  4. Twin prime conjecture. There are infinitely many primes pp such that p+2p+2 is also a prime.

    This is one of the central open problems about gaps between consecutive primes. It is equivalent to the assertion that infinitely many integers nn make both nn and n+2n+2 prime.

    source: Paweł Lewulis, “Almost primes in various settings”, arXiv:1806.09034 (2020).

  5. Let a twin prime be a pair of prime numbers differing by 22, equivalently primes pp and p+2p+2.

    Twin prime conjecture. There exist infinitely many twin primes.

    This is one of the classical open problems in number theory. The source notes that the conjecture remains open, while bounded-gap results show that infinitely many prime pairs occur with some fixed bounded difference; they do not establish the case of difference 22.

    source: Mbakiso Fix Mothebe, “Sieve methods and the twin prime conjecture”, arXiv:1909.02205 (2023).

  6. Let (pn)n∈N(p_n)_{n\in\mathbb{N}} denote the sequence of prime numbers in increasing order. Twin Prime Conjecture. There are infinitely many values n∈Nn\in\mathbb{N} such that

    pn+1−pn=2.p_{n+1}-p_n=2.

    This is the classical twin prime conjecture, asserting the existence of infinitely many pairs of primes differing by two. It remains unproved.

    source: Bryce Alan Christopherson, “Ramsey-Theoretic Characterizations of Classically Non-Ramseyian Problems”, arXiv:2502.04311 (2025).

  7. Let nn range over the positive integers. Twin prime conjecture. There are infinitely many pairs of primes of the form n,n+2n,n+2. This is the k=2k=2 case of the prime-tuple prediction represented by Dickson's conjecture and remains open.

    source: Andrew Darlington, “Hopf-Galois structures on separable field extensions of degree related to Cunningham chains”, arXiv:2508.03384 (2025).

  8. Let pp and p+2p+2 be a pair of twin primes, meaning that both numbers are prime. Twin primes conjecture. There are infinitely many pairs of twin primes p,p+2p,p+2. The conjecture is one of the central open problems in analytic number theory. Bounded-gap results show that infinitely many pairs of primes occur within a fixed distance, but it remains unknown whether infinitely many pairs differ by exactly 22.

    source: David Lowry-Duda, “A Friendly Intro to Sieves with a Look Towards Recent Progress on the Twin Primes Conjecture”, arXiv:1401.7555 (2014).

  9. Let pp range over prime numbers. Euclid's conjecture. There exist infinitely many prime pairs pp and p+2p+2, such as 3,53,5; 5,75,7; 11,1311,13; 17,1917,19; and so on. This is the twin prime conjecture, a classical open problem asserting that the gap 22 occurs infinitely often between consecutive primes.

    source: Guangchang Dong, “Infinitely many pairs of primes p and p+2”, arXiv:1406.4996 (2019).

References
  1. G. H. Hardy and J. E. Littlewood, "Some problems of 'Partitio numerorum'; III: On the expression of a number as a sum of primes", Acta Math. 44 (1923), 1–70. doi:10.1007/BF02403921
  2. J. R. Chen, "On the representation of a larger even integer as the sum of a prime and the product of at most two primes", Sci. Sinica 16 (1973); reprinted in The Goldbach Conjecture (ed. Y. Wang), World Scientific (2002), 275–294. doi:10.1142/9789812776600_0021
  3. D. A. Goldston, J. Pintz and C. Y. Yıldırım, "Primes in tuples I", Ann. of Math. 170 (2009), 819–862. doi:10.4007/annals.2009.170.819
  4. Y. Zhang, "Bounded gaps between primes", Ann. of Math. 179 (2014), 1121–1174. doi:10.4007/annals.2014.179.3.7
  5. J. Maynard, "Small gaps between primes", Ann. of Math. 181 (2015), 383–413. doi:10.4007/annals.2015.181.1.7
  6. D. H. J. Polymath, "Variants of the Selberg sieve, and bounded intervals containing many primes", Res. Math. Sci. 1 (2014), Art. 12. doi:10.1186/s40687-014-0012-7, arXiv:1407.4897
  7. J. Stadlmann, "On primes in arithmetic progressions and bounded gaps between many primes", Adv. Math. 468 (2025), 110190. doi:10.1016/j.aim.2025.110190, arXiv:2309.00425
  8. J. Stadlmann, "Bounded gaps between primes", preprint (2026). arXiv:2608.31126
  9. F. Charton, L. Hong, K. Lau, K. Ono, G. Remy, H. C. Siu, A. A. Swaminathan, J. Thorner and Y. Xie, "A new bound for small gaps between primes", preliminary draft (2026). primegaps.axiommath.ai/bgp212.pdf
  10. D. H. J. Polymath, "New equidistribution estimates of Zhang type", Algebra Number Theory 8 (2014), 2067–2199. doi:10.2140/ant.2014.8.2067
  11. Wikipedia, Brun's theorem.
  12. Wikipedia, Twin prime.

Source: Wikipedia

Progress summary

Refreshed
Open · progress reported

The twin prime conjecture remains open, although mathematicians have proved that infinitely many prime gaps are bounded by a small constant.

The conjecture, associated with de Polignac’s 1849 work, asserts that infinitely many primes pp have p+2p+2 prime. Equivalently, it asks whether H1=2H_1=2, where H1H_1 is the limiting smallest prime gap.

Known results

  • Chen, 1973: infinitely many primes pp have p+2p+2 prime or semiprime.
  • Zhang, 2013: infinitely many prime gaps satisfy H1≤7⋅107H_1\le 7\cdot 10^7.
  • Maynard and Tao, 2013–2015: new sieve methods established bounded gaps.
  • Polymath8b, 2014: the unconditional bound became H1≤246H_1\le246.

2026 bounded-gap improvements

Stadlmann’s August 31, 2026 preprint claims H1≤240H_1\le240. A preliminary 2026 draft by Charton, Hong, Lau, Ono, Remy, Siu, Swaminathan, Thorner and Xie claims H1≤212H_1\le212. These improve bounded-gap estimates but do not establish H1=2H_1=2; the newer claim is unverified.

Current status (as of November 2026): Infinitely many prime gaps of size at most 246246 are established, with newer bounds of 240240 and a claimed 212212 not independently verified; the twin prime conjecture remains open.

Sources

Solutions 0

No solutions have been posted yet.