OpenPath: Public Data Pathology Foundation Models and Leakage Free Evaluation Training, reproduction, and evaluation code. 🔗 GitHub · Checkpoints · Corpus OpenPath is a vision foundation model for computational pathology: a ViT g/14 encoder pre trained with self supervision ( DINOv2 + gram anchoring ) on public only whole slide histopathology tiles. This repository contains the training, reproduction, and evaluation code plus the released weight ( teacher checkpoint.pth = training 316250 ). The corpus and the full checkpoint set are hosted separately (see below). Headline result. On AMC HCC ST — a contamination free in house Asan Medical Center hepatocellular carcinoma spatial transcriptomics cohort, the least leakage prone benchmark since no public foundation model was trained on it — OpenPath ranks 1 among seven foundation models (mean Pearson: OpenPath 0.323 UNI2 h 0.301 OpenMidnight 0.300 Virchow2 0.292 prov gigapath 0.286 Phikon v2 0.274 UNI 0.257). Released checkpoint: training 316250 (in openpath checkpoints ). See Evaluation. Encoder: ViT g/14 (reg4), 1536 dim CLS embedding Objective: DINO + iBOT + KDE (DINOv2) with gram anchoring (technique from DINOv3, re implemented) Da…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy