whatmodel.sh / MedGemma 27B IT (multimodal)

MedGemma 27B IT (multimodal)

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

MedGemma 27B IT (multimodal) is an open-weights language model from Google, released July 2025. Its published size is 28.84B parameters. Licence: health-ai-developer-foundations.

Catalogue identity notes

google/medgemma-27b-it @2d3e00e (created 2025-07-09, licence health-ai-developer-foundations). Google MedGemma (multimodal 27B variant (July 2025).) built on Gemma 3; gated under the Health AI Developer Foundations terms. Independent rows: BRIDGE clinical zero-shot (87 tasks) and Open PT-LLM.

Model information sources

Release date ↗ · Weights licence ↗

HF safetensors metadata 28,842,036,848 parameters.

57.3±3.8#271 of 949 LLMsfrom 2 benchmark results across 2 sources* thinly tested, treat the rank as provisional

$ how it compares

The LLMs ranked either side of MedGemma 27B IT (multimodal), out of 1132 with a score, then more from Google.

#modelscorewhereRAMreleased
353Gemini 2.0 Flash Thinking Experimental (01-21) Google57±2comparison only2025-01
354DeepSeek R1 0528 Qwen3 8B DeepSeek57±1api $0.068/M +local~6.3GB2025-05
355MedGemma 27B IT (multimodal) Google57±4local~19GB2025-07
356Granite 4.2 3B IBM57±1api $0.052/M +local~3.7GB2026-08
357Command A (03-2025) Cohere57±1api $4.38/M +local~37GB2025-03
more from Google
9Gemini 4 Argon (limited preview) Google82±1limited preview2026-09
23Gemini 3.8 Flash Google77±1api $1.50/M2026-09
32Gemini 3.7 Flash Google76±1api $1.50/M2026-08

see all LLMs · find the ones that fit your machine

$ what it's good at

other tests2 not in a task · 94% of its score
testresultvs the others in its rundifficultycountswhy
BRIDGE clinical text, zero-shot average (87 tasks)40.8% ✎▁▁▁▁▁▂▂▃▄▆█▃ beats 79/97hard · avg 32%▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 70%blunt · at its level
Open PT-LLM Leaderboard average (9 Portuguese tasks)77.5% ✎▃▁▁▁▁▁▁▁▂█▅▁ beats 240/276medium · avg 73%▇▇▇▇▇▇▇▇▇ 23%at its level · noisy

$ where to run it

llama.cpp · recommended

version

  1. Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
    macOS / Linux
    curl -LsSf https://llama.app/install.sh | sh
    Windows
    irm https://llama.app/install.ps1 | iex

    other ways to install ↗

  2. Run MedGemma 27B IT (multimodal) · needs ~16.5 GB · from huggingface.co/unsloth
    chat in the terminal
    llama cli -hf unsloth/medgemma-27b-it-GGUF:Q4_K_M
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf unsloth/medgemma-27b-it-GGUF:Q4_K_M

LM Studio

version

  1. Install LM Studio · Open the app once before using the lms command.

    download it (macOS / Windows / Linux) ↗

  2. Run MedGemma 27B IT (multimodal) · needs ~16.5 GB · from huggingface.co/unsloth

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/unsloth/medgemma-27b-it-GGUF@Q4_K_M -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run MedGemma 27B IT (multimodal) · needs ~17 GB · from ollama.com
    chat in the terminal
    ollama run medgemma:27b

MLX (Mac)

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run MedGemma 27B IT (multimodal) · needs ~18.6 GB, Mac only · from huggingface.co/andosen
    chat in the terminal
    mlx_lm.chat --model andosen/medgemma-27b-it-mlx-5Bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model andosen/medgemma-27b-it-mlx-5Bit

Official weights and documentation: huggingface.co/google ↗

$ how the score was worked out · BRIDGE Medical Leaderboard (H… 70%, Open Portuguese LLM Leaderboa… 23%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
BRIDGE Medical Leaderboard (Harvard Medical School / Brigham and Women's YLab)1▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 70%MIT
Open Portuguese LLM Leaderboard (archived)1▇▇▇▇▇▇▇ 23%Apache-2.0
prior · pull toward the middle▇▇ 6%
weighted average= 57.3

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json