Dataset · MEDHEB
MedHEB-Bench medical benchmark sample
MEDHEB/MedHEB-Bench_Sample
A small test sample of the MedHEB-Bench benchmark for evaluating multimodal models.
What's inside
A self-contained test sample of the MedHEB-Bench benchmark, containing configurations for 2D, 3D, and text tasks with relative image paths. The dataset is designed for model validation and evaluation.
What you can do with it
- Benchmarking medical multimodal models
- Evaluating AI performance on 2D and 3D images
- Languages
- English
- Size
- up to 1K rows
- Downloads this month
- 30.5K
- Likes
- 1
How to load it
from datasets import load_dataset
ds = load_dataset("MEDHEB/MedHEB-Bench_Sample")