Model reference · open weights

RudolfV-2-B

RudolfV-2-B is an open-weight embedding model from Aignostics. RudolfV-2-B (FP32) weighs 171 MB; the smallest configuration that runs it is RTX 3060 12 GB.

• RudolfV-2-B is a vision transformer foundation model developed by Aignostics for computational pathology and image feature extraction.
• It utilizes a ViT-B/8 backbone with 85 million parameters and was distilled from the larger RudolfV 2 model.
• The model is licensed under CC BY-NC-ND 4.0 with supplementary terms restricting use to academic research and prohibiting clinical or commercial applications.

Summary of the Aignostics/RudolfV-2-B model card, 2026-10-01

Hugging Face ↗