The article introduces Prov-GigaPath, a whole-slide pathology foundation model designed to address the unique computational challenges of digital pathology. Prov-GigaPath is pre-trained on a large dataset, Prov-Path, which consists of 1.3 billion 256 × 256 pathology image tiles from 171,189 whole slides, covering more than 30,000 patients and 31 major tissue types. The model is trained using GigaPath, a novel vision transformer architecture that adapts the LongNet method to handle the large number of image tiles in gigapixel slides. Prov-GigaPath demonstrates state-of-the-art performance on various digital pathology tasks, including cancer subtyping and pathomics, with significant improvements over existing models. The article also explores the potential of Prov-GigaPath in vision-language pretraining by incorporating pathology reports, showing its capability in zero-shot subtyping and mutation prediction. Overall, Prov-GigaPath is an open-weight foundation model that leverages real-world data and whole-slide modeling to enhance clinical diagnostics and decision support in computational pathology.The article introduces Prov-GigaPath, a whole-slide pathology foundation model designed to address the unique computational challenges of digital pathology. Prov-GigaPath is pre-trained on a large dataset, Prov-Path, which consists of 1.3 billion 256 × 256 pathology image tiles from 171,189 whole slides, covering more than 30,000 patients and 31 major tissue types. The model is trained using GigaPath, a novel vision transformer architecture that adapts the LongNet method to handle the large number of image tiles in gigapixel slides. Prov-GigaPath demonstrates state-of-the-art performance on various digital pathology tasks, including cancer subtyping and pathomics, with significant improvements over existing models. The article also explores the potential of Prov-GigaPath in vision-language pretraining by incorporating pathology reports, showing its capability in zero-shot subtyping and mutation prediction. Overall, Prov-GigaPath is an open-weight foundation model that leverages real-world data and whole-slide modeling to enhance clinical diagnostics and decision support in computational pathology.