---

Gruslys and Letzter's Theorem 1.2 (arXiv:2008.05311, v1 of 12 August 2020, v2 of 14 August 2020 cited, p. 2) says that every $2$-coloring of the edges of $K_n$ admits $n^2/12+o(n^2)$ edge-disjoint monochromatic triangles. This is the conclusion Problem 76 asks about, in the paper's notation $n^2/12+o(n^2)$ for the site's $(1+o(1))n^2/12$, and the paper presents it as confirming its Conjecture 1.1, Erdős's "Problem 14" of 1997. The bound is best possible by the balanced two-part coloring, so the theorem determines the guaranteed count to first order. The proof passes to the fractional problem by the Haxell--Rödl transference (Corollary 2.2, p. 3) and proves the fractional extremal result Theorem 2.3 (p. 4): for $n\ge26$ every red-blue coloring of $K_n$ has a fractional monochromatic triangle packing of total edge weight at least $\lfloor(n-1)^2/4\rfloor$, with equality only when one color class is a balanced complete bipartite graph minus a matching (the printed statement says "the union of" it "with a matching"; the proof on pp. 17--18 concludes "minus a matching"). Two ingredients lie outside the paper: Theorem 2.11 (p. 7), proved in the companion preprint arXiv:2008.05313, and the computer-search certificates behind Lemma 2.8 (p. 6); neither is held here.

Acceptance. Reviewed: the site's curator, T. F. Bloom, labels the problem PROVED and credits the proof to this paper in the problem's commentary, which records that the answer is yes (page last edited 23 January 2026, accessed 2026-09-18); the discussion thread holds one comment correcting a misprinted name, nothing mathematical, and the proof-claim tab is empty. The curator is independent of the authors. Not refereed: on 2026-09-18 the arXiv record carried no journal reference, a Crossref bibliographic query for the title found no record, and the three citing works in the Semantic Scholar list were the companion paper and two papers on tournament inversions, so no refereed version is known. This page rests on the statements of Conjecture 1.1, Theorems 1.2, 1.3 and 2.3, Lemma 2.8 and Theorem 2.11, not on the proofs; nothing here is independent review.

Depends on. Nothing in this wiki; the result is the paper's own theorem, with the companion preprint and the computer certificates it cites.