Med-PaLM 2

// Overview

Med-PaLM 2 is a medical large language model from Google, built by fine-tuning the PaLM 2 model on medical data. It was the first AI system to reach expert level on USMLE-style questions from the MedQA benchmark, scoring 86.5%.

In a study of consumer medical questions, a panel of physicians preferred Med-PaLM 2's answers over physician-written answers across most of the axes the study measured. The model powers MedLM, a family of healthcare-tuned foundation models that Google offers to allowlisted Google Cloud customers through the Vertex AI platform.

// Benchmarks

1. MedQA (USMLE)86.5%

Scores on a 0–100 scale (25-point gridlines); higher is better. Each benchmark links to its published source.

// Strengths

• Expert-level accuracy on USMLE-style medical question answering
• Tuned specifically for medical and healthcare use
• Available as a managed API (MedLM) on Google Cloud Vertex AI

// Best for

• Answering medical exam-style and clinical knowledge questions
• Summarizing and drafting healthcare documentation within Google Cloud
• Powering healthcare applications built on Vertex AI

// Med-Gemini / Med-PaLM (medical) — every version

The full lineage of the Med-Gemini / Med-PaLM (medical) line, newest first. Every version has its own page — click any to compare specs, benchmarks and pricing.