Model reference · open weights

KRONOS2

Embeddings MahmoodLab Image embed 1 build Non-commercial 1k dl/mo

KRONOS2 is an open-weight embedding model from MahmoodLab. KRONOS2 (BF16) weighs 440 MB; the smallest configuration that runs it is RTX 3060 12 GB.

  • KRONOS2 is a marker-aware vision foundation model developed by MahmoodLab for extracting features from multiplex immunofluorescence spatial proteomics images.
  • It utilizes a DINOv2 ViT-B/16 architecture to generate 768-dimensional embeddings from 256 by 256 pixel patches, supporting a pretraining vocabulary of 268 markers.
  • The model is released under the CC-BY-NC-ND 4.0 license, restricting its use to non-commercial academic research, and requires gated access via an institutional email address.

Summary of the MahmoodLab/KRONOS2 model card, 2026-10-01

What it is

Released byMahmoodLab
Released2026-04-20
VRAM440 MB for the weights

What it runs on

Memory and cards for KRONOS2 (BF16)

440 MBweights, file size
1.1 GBruntime overhead
CardRunsMemory
RTX 3060 12 GBfits11.6 GB
RTX 4060 Ti 16 GBfits15.4 GB
RTX 3090 24 GBfits23.4 GB
RTX 4090 24 GBfits23.4 GB
RTX 5090 32 GBfits31.0 GB
L40S 48 GBfits44.0 GB
A100 80 GBfits78.2 GB
H100 80 GBfits78.1 GB
RTX PRO 6000 Blackwell 96 GBfits93.8 GB
DGX Spark (GB10) 128 GB unifiedfits107 GB
H200 141 GBfits138 GB
B200 180 GBfits176 GB

How it works

How embedding models work

Your textsentence / documentEncodermaps meaningVectorlist of numbersAn embedding model turns text into a vector, so similar meanings sit close together, which is the basis of search and RAG.
© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms