To accelerate the national transition to long-read genome sequencing (lrGS), we established a collaborative framework uniting two complementary initiatives spanning population genomics and rare disease diagnostics: (i) Genome of Sweden (GoS), a population-based project aligned with the Genome of Europe, and (ii) GMS-LRS, a prospective nationwide clinical study of newly referred children and adults with rare neurological disorders. In each project, over 1000 individuals are sequenced using PacBio Revio and the datasets are processed in Nallo, a community driven lrGS analysis pipeline. The first GoS data release includes lrGS from 480 individuals (mean read length 14 kb; 22.2× coverage; ~4.6 million SNVs; 27k SVs per sample), enabling diploid de novo assemblies and a CpG 5mC reference dataset. In parallel, GMS-LRS has enrolled 211 probands and sequenced 310 samples (160 singletons and 51 trios), referred from all Swedish university healthcare regions. Overall, 62% of probands were children, and the most common indications were neurodevelopmental disorders (42%) and neuromuscular disease (24%). GoS-derived population allele frequencies were continuously integrated through a series of early data releases, supporting variant filtering and interpretation in the clinical cohort. In an initial subset of cases, lrGS achieved a diagnostic yield of ~38%, identifying likely pathogenic/pathogenic variants across multiple genes. Detected findings spanned aneuploidy, STR expansions, SNV/indels, and structural variants. Together, GoS and GMS-LRS form a national framework that integrates population reference data with clinically validated implementation, supports the transition from short-read to long-read genome sequencing in rare disease diagnostics, and will enable future diagnostic advances through methylation and episignature analysis.