Medical LLM - Medium

This model is an advanced clinical decision-support system built for structured medical reasoning rather than simple knowledge retrieval. It analyzes symptoms, diagnostics, and longitudinal patient histories to guide complex diagnostic and treatment decisions in line with established clinical guidelines. With multimodal vision capabilities, it can also interpret medical images and visual documents alongside text, broadening the range of clinical inputs it can reason over. Now grounded in curated, specialty-specific medical knowledge bases, it anchors conclusions to vetted clinical evidence for greater factual reliability. It delivers transparent, step-aware reasoning, weighs competing hypotheses, and communicates uncertainty to support risk-aware clinical judgment. Benchmarking Results: - Achieves 94.5% average across OpenMed benchmarks; - Medical genetics: 99%; professional medicine: 98%; - Clinical knowledge comprehension: 95%; college biology mastery: 97.5%; - MedHallu: 97.5%; Anatomy: 93% - Achieves 94.7% on MedQA and 84% on PubMedQA; - Safety & reliability: 95.5% on hallucination detection, inappropriate-content , refusal of unethical/illegal requests; - Fairness & bias: 98% pass rate on name bias, 94% on racial bias , and 94.67% on anchoring-bias resistance Recommended Instance type for this model: Standard_NC96ads_A100_v4