HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks

medgemma-27b-text-it

google/medgemma-27b-text-it

kind
original
quant
bf16 (16-bit)
family
gemma
modality
text-generation
root model
google/gemma-3-27b-pt
provider
google
license
other
language
—
context
—
library
transformers
vRAM (GB)
—
size on disk
—
downloads
30,027
likes
458
created
—
last modified
2025-09-16T23:34:17+00:00 (2025)
huggingface
https://huggingface.co/google/medgemma-27b-text-it
tags
transformerssafetensorsgemma3_texttext-generationmedicalclinical-reasoningthinkingconversationalarxiv:2507.05201arxiv:2501.19393arxiv:2303.15343arxiv:2009.13081arxiv:2102.09542arxiv:2411.15640arxiv:2404.05590arxiv:2501.18362base_model:google/gemma-3-27b-ptbase_model:finetune:google/gemma-3-27b-ptlicense:othertext-generation-inferenceendpoints_compatibleregion:us

Benchmark scores (2)

benchmarkcategoryscorecontextconfevidence
MMLUknowledge87.0MMLU Med (text only)medium| MMLU Med (text only) | 87.0 | 83.3 | 70.0 | 67.2 |
MedQAdomain74.9MedQA (4-op)medium| MedQA (4-op) | 89.8 (best-of-5) 87.7 (0-shot) | 74.9 | 64.4 | 50.7 |

Discussions (0)

none scraped.

Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.