Open Medical LLM - MMLU Medical Genetics: leaderboard

MMLU Medical Genetics subset: questions on hereditary diseases, genetic testing, chromosomal abnormalities, and molecular genetics.

Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 193 models tracked.

Top models

#ModelScoreOverall rank
1GPT-491#438
2Llama-3-8B-Instruct-abliterated-dpomix85#1049
3Daredevil-8B-abliterated84#846
4Meta-Llama-3-8B-Instruct-abliterated-v384#1208
5Daredevil-8B-abliterated-dpomix84#838
6Llama 3 8B Ita83#926
7Llama-3-SauerkrautLM-8B-Instruct83#895
8Barcenas-Llama3-8B-ORPO83#949
9NeuralLLaMa-3-8B-DT-v0.183#923
10llama-3-merged-linear83#887
11ChimeraLlama-3-8B-v383#1055
12Llama 3 8B Instruct82#1129
13NeuralLLaMa-3-8B-ORPO-v0.382#1174
14Yi-1.5-9B81#1241
15suzume-llama-3-8B-multilingual81#1084

No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=open-medical-llm-mmlu-medical-genetics · How It Works · Data refreshed daily, snapshot 2026-10-20.