Hulu-Med-4B: benchmark results

Provider: Other.

Aggregate Rating 1501 ± 24, rank #2344 of 10916 rated models, from 45 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
MedStreamBench - Future43.27Content correctness (%; questions about events after the que87.5
Med-R2 - Lesion Level57.9Accuracy (%) on the Med-R2 Lesion Level task (detecting and 84.6
MedStreamBench - Future Responsiveness14.56Responsiveness (%; timing of the first non-unanswerable answ75
MedStreamBench - Proactive Responsiveness20.88Responsiveness (%; timing of the first alert relative to the75
ADRD-Bench - Unified QA84.35Exact-match accuracy (0-1, shown times 100) on the 1,438 Alz51.4
Med-R2 - Anatomy Level60.3Accuracy (%) on the Med-R2 Anatomy Level task (locating and 46.2
Med-R2 - Joint QA58.4Accuracy (%) on the Med-R2 Joint QA task (answering all reas46.2
KorMedMCQA-V - Non-Stress Test46.4Accuracy (%) on the questions carrying the 54 fetal non-stre46
KorMedMCQA-V - MRI54.3Accuracy (%) on the questions carrying the 40 MRI images (Ko45
KorMedMCQA-V - Ultrasound45.6Accuracy (%) on the questions carrying the 138 ultrasound im44
KorMedMCQA-V - X-Ray45.4Accuracy (%) on the questions carrying the 586 X-ray images 42
KorMedMCQA-V - Peripheral Blood Smear48.1Accuracy (%) on the questions carrying the 49 peripheral blo40

Interactive version: theaggregate.ai/model?slug=hulu-med-4b · How It Works · Data refreshed daily, snapshot 2026-11-08.