Quasi-cliques in inhomogeneous random graphs

Given a graph $G$ and a constant $γ\in [0,1]$, let $ω^{(γ)}(G)$ be the largest integer $r$ such that there exists an $r$-vertex subgraph of $G$ containing at least $γ\binom{r}{2}$ edges. It was recently shown that $ω^{(γ)}(G)$ is highly concentrated when $G$ is an Erdős-Rényi random graph (Balister, Bollobás, Sahasrabudhe, Veremyev, 2019). This paper provides a simple method to extend that result to a setting of inhomogeneous random graphs, showing that $ω^{(γ)}(G)$ remains concentrated on a small range of values even if $G$ is an inhomogeneous random graph. Furthermore, we give an explicit expression for $ω^{(γ)}(G)$ and show that it depends primarily on the largest edge probability of the graph $G$.