Medmarks: An Open-Source LLM Benchmark Suite for Medical Tasks

Benjamin Warner, Ratna Sagari Grandhi, Max Kieffer, Aymane Ouraq, Saurav Panigrahi, Geetu Ambwani, Kunal Bagga, Nikhil Khandekar, Arya Hariharan, Nishant Mishra, Manish Ram, Shamus Sim Zi Yang, Ahmed Essouaied, Adepoju Jeremiah Moyondafoluwa, Robert Scholz, Bofeng Huang, Molly Beavers, Srishti Gureja, Anish Mahishi, Sameed Khan et al. (15 additional authors not shown)

Keywords: Medical AI, Large Language Models, Benchmarking, Model Evaluation, Health NLP, Open-Source Datasets, LLM-as-a-Judge

TL;DR: We introduce Medmarks, a fully open-source LLM evaluation suite for medical tasks with 30 benchmarks and evaluated 61 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 \textsc{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). Code is available at \url{https://anonymous.4open.science/r/med-lm-envs-B76C/}.

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 113