What are the strengths of this model?

Specialized optimization for medical field Wide 128K context window High reliability from DeepMind

What are the weaknesses of this model?

Closed license system High computational resource demands Adaptability to general tasks

What are the best use cases?

Advanced analysis of medical information Specialized medical chat Processing large-scale medical literature

Back to Models

DeepMindProprietary

MedGemma 27B Instruct

Name: MedGemma 27B Instruct
Author: DeepMind

MedGemma 27B Instruct is a foundation model specialized in the medical field, developed by DeepMind. It has a scale of 27 billion parameters and is a chat model that supports a long context window of up to 128K.

Parameters

270.0B

Context Window

128K

License

https://creativecommons.org/licenses/by-sa/4.0/

Release Date

2025-07-09

API Pricing

API pricing for this model is not yet available

Strengths

・Specialized optimization for medical field
・Wide 128K context window
・High reliability from DeepMind

Weaknesses

・Closed license system
・High computational resource demands
・Adaptability to general tasks

Use Cases

・Advanced analysis of medical information
・Specialized medical chat
・Processing large-scale medical literature

Deep Analysis

Parameters

27B

Available in text-only and multimodal variants

MedQA

87.7%

Within 3 points of DeepSeek R1 at ~1/10th the inference cost

Chest X-ray Reports

81% clinically sufficient

US board-certified radiologist review for 4B variant

Base Model

Gemma 3

Medically optimized variant of Gemma 3

Release Date

July 9, 2025

Part of Health AI Developer Foundations (HAI-DEF)

License

HAI-DEF Terms of Use

Special health AI license, not standard Gemma

Deployment

Single GPU

Can run on a single GPU; 4B variant runs on mobile

Modalities

Text + Images

Multimodal: medical text and imaging (X-rays, dermatology, ophthalmology, histopathology)

Strengths

・Best-in-class MedQA (87.7%) among models under 50B parameters
・Within 3 points of DeepSeek R1 at 1/10th the inference cost
・Multimodal: handles medical images (X-rays, dermatology, ophthalmology, histopathology)
・Retains general Gemma capabilities (instruction-following, non-English languages)
・Specialized SigLIP image encoder pre-trained on medical data
・Runs on a single GPU

Weaknesses

・Not a general-purpose model - specialized for medical tasks
・HAI-DEF license more restrictive than Apache 2.0
・Requires fine-tuning for optimal performance on specific medical tasks
・Limited to text and image output (no audio/video)
・May not be suitable for direct clinical use without validation

Competitor Comparison

Model	Arena	SWE	GPQA	Price
MedGemma 27B	N/A	N/A	87.7% (MedQA)	Free (open weights)
DeepSeek R1	~1470	~70%	~90% (MedQA)	Free (open weights)
BioMistral 7B	N/A	N/A	~60% (MedQA)	Free (open weights)
GPT-5 (general)	~1480	~80%	~85% (MedQA est)	$5/$20 per 1M

Overview

MedGemma 27B is Google's most capable open medical model, part of the Health AI Developer Foundations (HAI-DEF) collection. Available in text-only and multimodal variants, it achieves 87.7% on MedQA - within 3 points of DeepSeek R1 at approximately 1/10th the inference cost. Specialized for medical text comprehension, report generation, and medical image interpretation.

Benchmarks & Performance

MedQA: 87.7% (among best under 50B). Competitive with larger models on medical retrieval and EHR interpretation. The 4B variant (separate model) achieved 64.4% MedQA and 81% clinically sufficient chest X-ray reports. After fine-tuning, achieves RadGraph F1 of 30.3 on chest X-ray report generation.

Detailed Comparison

Near DeepSeek R1 medical performance at dramatically lower cost. Much smaller than general-purpose frontier models but superior on medical tasks within its size class. The 27B text variant is the strongest; multimodal variant adds image understanding.

Community Feedback

Well-received by healthcare AI developers. The open nature enables privacy-preserving deployment. Used for medical report generation, clinical decision support, and medical image analysis. Researchers appreciate the specialized SigLIP encoder for medical imaging.

Use Cases

Medical text reasoning, clinical report generation, medical image interpretation (X-rays, dermatology, ophthalmology, histopathology), EHR data analysis, medical question answering. Designed as a starting point for healthcare AI development with fine-tuning for specific clinical tasks.

Latest News

Released July 9, 2025. 27B multimodal variant announced as complement to existing 4B multimodal and 27B text-only models. MedSigLIP encoder also released separately for classification/retrieval tasks.

Sources

Analysis generated: 2026-05-24