TRACE · Foundation model
A vision encoder for toxicologic pathology.
The model beneath everything Tremont AI builds. A representation of tissue learned at very large scale on preclinical data.
TRACE embedding space
Kidney · SD rat · H&E · 20×
What it is
The vision encoder every other model runs on.
A whole-slide image is a gigapixel of tissue. TRACE turns it into a representation the rest of the stack can work with, one that keeps what a pathologist would look at and drops what they wouldn’t.
Research & partnerships
Build on the foundation.
We collaborate with research partners on what a toxicologic-pathology encoder can do.
Start a conversation →