Strong gravitational lensing has the potential to provide a powerful probe of astrophysics and cosmology, but fewer than 1000 strong lenses have been confirmed previously. With \ang{;;0.16} resolution covering a third of the sky, the \Euclid telescope will revolutionise strong lens finding, with \num{170000} lenses forecasted to be discovered amongst its 1.5 billion galaxies. We present an analysis of the performance of five machine-learning models at finding strong gravitational lenses in the quick release of \Euclid data (Q1), covering 63\,deg$^{2}$. The models are validated with citizen scientists and expert visual inspection. We focus on the best performing network: a fine-tuned version of the \texttt{Zoobot} pretrained model, originally trained to classify galaxy morphologies in heterogeneous astronomical imaging surveys. Of the one million Q1 objects that \texttt{Zoobot} was tasked to find strong lenses within, the top 1000 ranked objects contained 122 grade A lenses (almost certain lenses), and 41 grade B lenses (probable lenses). A deeper search with the five networks combined with visual inspection discovered 250 (247) grade A (B) lenses, of which 224 (182) are ranked in the top \num{20000} by \texttt{Zoobot}. When extrapolated to the full \Euclid survey, the highest ranked one million images will contain \num{75000} grade A or B strong gravitational lenses.
Comment: Paper submitted as part of the A&A Special Issue `Euclid Quick Data Release (Q1)', 22 pages, 14 figures