multimodal benchmark
MedXpertQA Leaderboard
MedXpertQA is a benchmark of expert-level medical knowledge and advanced reasoning with 4,460 questions spanning 17 specialties and 11 body systems, including text-only and multimodal subsets.
Updated Sep 5, 2026
Models12
Model coverage12
MetricScore
EvidenceB
MedXpertQA Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelME | Score78.40% | Percentile100.00% | Participants12 | EvidenceC | Evaluated |
| Rank02 | ModelAC | Score67.30% | Percentile90.91% | Participants12 | EvidenceC | Evaluated |
| Rank03 | ModelAC | Score62.40% | Percentile81.82% | Participants12 | EvidenceC | Evaluated |
| Rank04 | ModelAC | Score61.40% | Percentile72.73% | Participants12 | EvidenceC | Evaluated |
| Rank05 | ModelGO | Score61.30% | Percentile63.64% | Participants12 | EvidenceC | Evaluated |
| Rank06 | ModelGO | Score58.10% | Percentile54.55% | Participants12 | EvidenceC | Evaluated |
| Rank07 | ModelGO | Score49.00% | Percentile45.45% | Participants12 | EvidenceC | Evaluated |
| Rank08 | ModelGO | Score48.70% | Percentile36.36% | Participants12 | EvidenceC | Evaluated |
| Rank09 | ModelMI | Score43.00% | Percentile27.27% | Participants12 | EvidenceC | Evaluated |
| Rank10 | ModelGO | Score28.70% | Percentile18.18% | Participants12 | EvidenceC | Evaluated |
| Rank11 | ModelGO | Score23.50% | Percentile9.09% | Participants12 | EvidenceC | Evaluated |
| Rank12 | ModelGO | Score18.80% | Percentile0.00% | Participants12 | EvidenceC | Evaluated |
MedXpertQA Highlights
The leading models and scores on this benchmark.
MedXpertQA Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for MedXpertQA
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is MedXpertQA?
What MedXpertQA measures and how its scores work.
MedXpertQA is a medical benchmark featuring expert-level exam questions with diverse medical images and clinical information.
It measures expert-level medical knowledge and advanced reasoning using Score, reported as a ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MedXpertQA
- Modality
- multimodal
- Primary category
- multimodal
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about MedXpertQA.
Which model scores highest on MedXpertQA?
Muse Spark is currently ranked first with 78.40%.
What are the top three models on MedXpertQA?
The current leaders are Muse Spark (78.40%), Qwen3.5-122B-A10B (67.30%), and Qwen3.5-27B (62.40%).
Which MedXpertQA model has the lowest official input price?
Qwen3.5-35B-A3B has the lowest matched official input price at $0.25 input / $2.0 output per 1M tokens.
Which models are fastest among MedXpertQA results?
The fastest matched records are Gemma 4 31B (22.86 tok/s via FriendliAI).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.