medgemma-4b-it
google · google/medgemma-4b-it
Weights license: Publisher terms: Health AI Developer Foundations License · code and components may use other terms
Quality
No comparable score
We track no public, comparable benchmark that covers this model yet. No score is inferred.
Reported by the publisher
- MIMIC CXR - macro F1 for top 5 conditions
- 88.9Imaging evaluations · Medical image classification
- CheXpert CXR - macro F1 for top 5 conditions
- 48.1Imaging evaluations · Medical image classification
- CXR14 - macro F1 for 3 conditions
- 50.1Imaging evaluations · Medical image classification
- PathMCQA - Accuracy (histopathology, internal)
- 69.8Imaging evaluations · Medical image classification
- US-DermMCQA - Accuracy
- 71.8Imaging evaluations · Medical image classification
- EyePACS - Accuracy (fundus, internal)
- 64.9Imaging evaluations · Medical image classification
- SLAKE - Tokenized F1 (radiology)
- 72.3Imaging evaluations · Visual question answering
- VQA-RAD - Tokenized F1 (radiology)
- 49.9Imaging evaluations · Visual question answering
Every reported result
- MedXpertQA - Accuracy (text + multimodal questions)
- 18.8Imaging evaluations · Knowledge and reasoning
- MedQA (4-op)
- 64.4Text evaluations
- MedMCQA
- 55.7Text evaluations
- PubMedQA
- 73.4Text evaluations
- MMLU Med
- 70.0Text evaluations
- MedXpertQA (text only)
- 14.2Text evaluations
- AfriMed-QA (25 question test set)
- 52.0Text evaluations
- EHRQA
- 67.6Medical record evaluations
Values as printed on the card, with the publisher's own settings (card section: Model information › Key performance metrics › Imaging evaluations; Model information › Key performance metrics › Text evaluations); column “MedGemma 4B”. Not comparable across publishers and not used in any ranking. Model card
Runs on
Fast memory holds it 118
The planning estimate fits the memory the model can use quickly: graphics card VRAM, or unified memory within the macOS GPU share.
Show all 118
Within memory, outside the fast part 1
Within the machine's memory, outside its fast part: on a graphics card the layers that do not fit run on the CPU from PC RAM (the listed RAM, or 32 GB assumed); on a Mac, above the macOS default GPU share. Not a speed figure.
Hardware profiles grouped by the Q4 planning estimate on each machine (dedicated GPUs reserve 2 GB; unified and system memory 25%, 2 to 8 GB): the fast memory holds it, it is within memory but outside the fast part (layers offloaded to PC RAM, or above the macOS GPU share), or the machine is one memory tier short. Not a measured fit or speed.
Measured speed
No published measurement for this checkpoint
Tokens per second are shown only when a source publishes them with hardware, build and setup. They are never estimated.
How to run
Runtime compatibility and tokens per second have not been measured for this entry. Published metadata and a memory estimate do not prove a successful local run.
Library in source: transformers
Documented ways to run it
The model card could not be read (missing or gated)
Technical details
What the source and the catalog store for this checkpoint. Missing fields are left out, never zero.
- Checkpoint / source
- google/medgemma-4b-it
- Task
- image-text-to-text
- Architecture
- Gemma3ForConditionalGeneration
- Library in source
- transformers
- Weight license
- Publisher terms: Health AI Developer Foundations LicenseGoogle-Health/medgemma README, License: the model is licensed under the Health AI Developer Foundations License · read 2026-10-18source
- Commercial use
- Allowed Health AI Developer Foundations Terms of Use (verbatim copy in the ScanCode LicenseDB, homepage developers.google.com/health-ai-developer-foundations/terms) · read 2026-10-18license text
Every stored field
- Revision checked
- 290cda5eeccbee130f987c4ad74a59ae6f196408
- Architecture
- Gemma3ForConditionalGeneration
- Library
- transformers
- Hub task tag
- image-text-to-text
- Declared base models
- google/medgemma-4b-pt
- Formats
- transformers, safetensors
- Access
- Gated (auto)
- Publisher kind
- Official
- Repository created
- 2025-05-19creation is not the model's release
- Repository updated
- 2025-10-28a commit is not a release date
- Hub downloads (30 days) · likes
- 415,640 · 1,117this repository only · read 2026-10-25
- Read from the Hub page
- transformers class, task tagsHub overview and file listing · read 2026-10-18
- Catalog state
- Hub config or weights found; no inventory benchmark link
Sources
Identity and published specifications come from the checkpoint page. Benchmark scores belong to the named release and do not measure speed on your computer.
Valid source snapshot: 2026-09-26. This source record has not received a complete technical review. Missing data remain unknown; a repository commit is not a model release date.






























































