models datasets
2 datasets tagged "models"
How Accurate Are 10 LLMs in 13,500 Triage Tests?
Emergency triage results compare 10 LLMs across 45 vignettes, three prompt strategies and 10 rounds, covering EM, NE and SE.
Top AI Models Compared: Coding & Reasoning Scores (2025)
Coding and reasoning scores are strongly positively correlated (r²=0.90), showing how closely model performance tracks across both benchmarks.