meeTARA MedGemma 4B IT (GGUF Multimodal Distribution)
meeTARA = Mental & Emotional Empathetic Technology for Adaptive Response & Assistance
This repository contains the quantized GGUF export of google/medgemma-4b-it with embedded meeTARA system guidance and multimodal vision projector support (mmproj).
Files in this Repository
| File | Description | Size |
|---|---|---|
meetara-medgemma-4b-it-Q4_K_M.gguf |
Primary 4-bit Quantized Medical LLM | 2.32 GB |
meetara-vl-medgemma-4b-it.gguf |
Multimodal Vision Projector (mmproj) |
0.79 GB |
How to Run
1. Multimodal Mode (Text + Medical Images / X-Rays)
To analyze medical images (e.g. X-rays, CT scans, skin lesions) alongside text prompts, load both the LLM and the vision projector:
Using llama.cpp (llama-mtmd-cli):
./llama-mtmd-cli \
-m meetara-medgemma-4b-it-Q4_K_M.gguf \
--mmproj meetara-vl-medgemma-4b-it.gguf \
--image "path/to/xray.jpg" \
-p "Describe the findings in this X-ray."
Using llama-cpp-python:
from llama_cpp import Llama
llm = Llama(
model_path="meetara-medgemma-4b-it-Q4_K_M.gguf",
chat_handler=your_mmproj_handler,
n_ctx=4096,
)
2. Text-Only Medical Q&A Mode
For text-based medical inquiries, health education, and symptom explanation:
./llama-cli \
-m meetara-medgemma-4b-it-Q4_K_M.gguf \
-p "What are general self-care steps for a mild tension headache?"
Medical Disclaimer & Guidelines
- Educational & Support Use Only: This AI model provides general health information and image interpretation summaries for educational and clinical decision support purposes.
- Not a Diagnostic Tool: It is not a replacement for a licensed healthcare provider or board-certified radiologist.
- Emergency Situations: For urgent medical emergencies, immediately contact local emergency services.
Credits & License
- Base Model: Google DeepMind (
google/medgemma-4b-it) - GGUF Conversion & meeTARA Packaging: meeTARA Project
- License: Apache 2.0 / Health AI Terms