The InclusionMed Collaboration · Medical AI Benchmark

InclusionMed

InclusionMed is an open medical AI evaluation initiative that organises benchmarks across three healthcare roles and three AI capabilities in an Atlas, currently containing five benchmarks including a private diagnostic reasoning benchmark, InclusionMed-DxFiles. The authors evaluated frontier large language models and aggregated results into the InclusionMed Index, with overall scores ranging from 17.5 to 49.1 out of 100. They report that the ten top-ranked models all scored higher on patient communication than on clinical decision-making, and that models with similar Index scores differed substantially in estimated inference cost.

Open the source ↗

Similar Research & Safety releases

Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.