Medmarks: A Comprehensive Open-Source LLM Benchmark Suite for Medical Tasks

TL;DR

Medmarks introduces an open-source, comprehensive benchmark suite for evaluating large language models on diverse medical tasks, addressing limitations of existing benchmarks.

Contribution

It provides a new open-source evaluation suite with 30 benchmarks, systematic evaluation of 61 models, and insights into model performance and biases in medical AI.

Findings

Frontier reasoning models outperform others across benchmarks.

Medically fine-tuned models outperform generalist models.

Models are susceptible to answer-order bias, especially smaller models.

Abstract

Evaluating large language models (LLMs) for medical applications remains challenging due to benchmark saturation, limited data accessibility, and insufficient coverage of relevant tasks. Existing suites have either saturated, heavily depend on restricted datasets, or lack comprehensive model coverage. We introduce Medmarks, a fully open-source evaluation suite with 30 benchmarks spanning question answering, information extraction, medical calculations, and open-ended clinical reasoning. We perform a systematic evaluation of 61 models across 71 configurations using verifiable metrics and LLM-as-a-Judge. Our results show that frontier reasoning models (Gemini 3 Pro Preview, GPT-5.1, & GPT-5.2) achieve the highest performance across both benchmarks, most frontier proprietary models are significantly more token efficient than open-weight alternatives, medically fine-tuned models outperform their generalist counterparts, and that models are susceptible to answer-order bias (particularly smaller models and Grok 4). A subset of our evals (Medmarks-T) can be directly used as reinforcement learning environments to post-train LLMs for medical reasoning. Code is available at https://github.com/MedARC-AI/Medmarks