Instructions to use Jesteban247/brats_medgemma-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Jesteban247/brats_medgemma-GGUF with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("visual-question-answering", model="Jesteban247/brats_medgemma-GGUF")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Jesteban247/brats_medgemma-GGUF", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use Jesteban247/brats_medgemma-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Jesteban247/brats_medgemma-GGUF:Q5_K_M
# Run inference directly in the terminal:
llama cli -hf Jesteban247/brats_medgemma-GGUF:Q5_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf Jesteban247/brats_medgemma-GGUF:Q5_K_M
# Run inference directly in the terminal:
llama cli -hf Jesteban247/brats_medgemma-GGUF:Q5_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf Jesteban247/brats_medgemma-GGUF:Q5_K_M
# Run inference directly in the terminal:
./llama-cli -hf Jesteban247/brats_medgemma-GGUF:Q5_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf Jesteban247/brats_medgemma-GGUF:Q5_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf Jesteban247/brats_medgemma-GGUF:Q5_K_M

Use Docker

docker model run hf.co/Jesteban247/brats_medgemma-GGUF:Q5_K_M

LM Studio
Jan
Ollama
How to use Jesteban247/brats_medgemma-GGUF with Ollama:
```
ollama run hf.co/Jesteban247/brats_medgemma-GGUF:Q5_K_M
```

Unsloth Studio

How to use Jesteban247/brats_medgemma-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Jesteban247/brats_medgemma-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for Jesteban247/brats_medgemma-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for Jesteban247/brats_medgemma-GGUF to start chatting

Docker Model Runner
How to use Jesteban247/brats_medgemma-GGUF with Docker Model Runner:
```
docker model run hf.co/Jesteban247/brats_medgemma-GGUF:Q5_K_M
```

Lemonade

How to use Jesteban247/brats_medgemma-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull Jesteban247/brats_medgemma-GGUF:Q5_K_M

Run and chat with the model

lemonade run user.brats_medgemma-GGUF-Q5_K_M

List all available models

lemonade list

Atomic Chat

🧠 Jesteban247/brats_medgemma-GGUF

This repository provides GGUF-format variants of the fine-tuned model
Jesteban247/brats_medgemma,
which originates from the BraTS (Brain Tumor Segmentation and TextBraTS) domain adaptation experiments detailed in
Jesteban247/medgemma-brats-experiments.

🧩 Model Description

These files are direct GGUF conversions of the brats_medgemma model, compatible with llama.cpp and the GGUF-my-repo conversion space.

The GGUF format allows efficient inference across CPU and GPU environments, offering a lightweight deployment option for medical visual-language applications such as MRI interpretation and radiology text reasoning.

🧮 Available Files

File	Precision	Size	Description
`model_f16.gguf`	FP16	7.77 GB	Full-precision float16 version
`model_q8.gguf`	Q8_0	4.13 GB	Balanced quantization for accuracy and speed
`brats_medgemma-q5_k_m.gguf`	Q5_K_M	2.83 GB	Medium quantization for balanced size and performance
`mmproj_model_f16.gguf`	FP16	851 MB	Multimodal projection weights (image-text alignment)