YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.


library_name: transformers license: apache-2.0 language: - bn - en base_model: - google/gemma-3-4b-it pipeline_tag: image-to-text tags: - medical - vision - gemma-3 - bangladesh

MedGemma Workflow AI

MedGemma Workflow AI is a specialized multimodal vision-language model designed for medical reasoning and image-to-text tasks. It bridges the gap between medical image processing and descriptive AI analysis, optimized for both English and Bengali contexts.

Model Details

Model Description

This model is a fine-tuned version of Google's Gemma-3-4b-it, specifically adapted to handle medical workflows. It is designed to interpret medical visual data (like X-rays or clinical images) and provide structured textual insights. As an aspiring AI developer and cybersecurity researcher from Bangladesh, I developed this model to explore the intersection of healthcare AI and secure model deployment.

  • Developed by: opdejoy
  • Model type: Multimodal Vision-Language Model
  • Language(s) (NLP): Bengali (bn), English (en)
  • License: Apache-2.0
  • Finetuned from model: google/gemma-3-4b-it

Uses

Direct Use

  • Medical image description and analysis.
  • Assisting researchers in medical document interpretation.
  • Visual Question Answering (VQA) in a medical context.

Out-of-Scope Use

  • Critical Medical Diagnosis: This model is for research and educational purposes. It should NOT be used as a primary tool for diagnosing patients.
  • Malicious Use: Any use involving the generation of fake medical reportsf
Downloads last month
8
Safetensors
Model size
4B params
Tensor type
F32
BF16
U8
Inference Providers NEW
This model isn't deployed by any Inference Provider. 馃檵 Ask for provider support