Easily compute clip embeddings and build a clip retrieval system with them

Fast, accurate & comprehensive text measurement & layout

AI agents running research on single-GPU nanochat training automatically

Perfect Green Screen Keys

Official Implementation of "Maximum Likelihood Reinforcement Learning (MaxRL)"

Gpu accelerated vahadane stain normalization for Digital Pathology workflows.

Hierarchical Image Pyramid Transformer - CVPR 2022 (Oral)

Stain normalization tools for histological analysis and computational pathology

SEED-Voken: A Series of Powerful Visual Tokenizers

[ICML 2025 Tokshop] One-D-Piece: Image Tokenizer Meets Quality-Controllable Compression

[ICLR26] Code and Pretrained Models for Brain-Semantoks - Brain Foundation Model

SurGen: 1020 H&E-stained Whole Slide Images With Survival and Genetic Markers

[ICLR 2026] PixNerd: Pixel Neural Field Diffusion

Pixio: a capable vision encoder dedicated to dense prediction, simply by pixel reconstruction

automatically create cortical flatmaps from FreeSurfer surfaces

[NeurIPS25 D&B Spotlight] A tile-level histopathology image understanding benchmark

Standardized benchmark for computational pathology foundation models.

A debugging and profiling tool that can trace and visualize python code execution

The best ChatGPT that $100 can buy.

PLM-ICD: Automatic ICD Coding with Pretrained Language Models

CogStack NLP, including the Medical Concept Annotation Tool MedCAT

A toolkit for developing foundation models using Electronic Health Record (EHR) data.

Official code of Franca: Nested Matryoshka Clustering for Scalable Visual Representation Learning

Marketplace ML experiment - training without backprop

H&E tailored Randaugment: automatic data augmentation policy selection for H&E-stained histopathology.

Reference PyTorch implementation and models for DINOv3

transform an entire GitHub repo into a simple text file so that you can upload it to an LLM (Claude, ChatGPT, etc.)