# ZJUI-AI4H/Hulu-Med

A Transparent Generalist Model towards Holistic Medical Vision-Language Understanding

Repository: https://github.com/ZJUI-AI4H/Hulu-Med
Canonical: https://ross.abutalabs.com/products/hulu-med
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2026-08-30T12:18:18+00:00

## Health v2 (maintenance only)
Score: 62/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 100, release rhythm 35, longevity 23
- inputs: {"age_days": 329, "days_push": 3, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1001, forks 36 (observed 2026-09-03T02:15:08.692171+00:00)

## What it is
Hulu-Med is a family of open-source transparent generalist medical vision-language models ranging from 4B to 235B parameters, covering text, 2D, 3D, and video medical modalities. It includes model weights on HuggingFace, inference support via vLLM, a Gradio demo, and the MedUniEval evaluation toolkit.

## Use cases
- run a medical vision-language model for radiology report understanding
- analyze medical images like CT scans with an AI model
- build a medical chatbot that understands clinical images and text
- evaluate medical multimodal models on benchmarks
- deploy a medical LLM locally with vLLM
- fine-tune or study open medical AI models
- answer health questions with an open-source medical model

## When to choose
- you need an open, transparent medical multimodal model with permissive Apache-2.0 licensing
- you want to handle diverse medical modalities including 2D, 3D, and video
- you need a range of model sizes from lightweight 4B to large MoE variants
- you want fast inference with vLLM and tensor parallelism support

## When to avoid
- you need a clinically certified or regulated medical device solution
- you lack GPU hardware for running large vision-language models
- you need a general-purpose (non-medical) assistant rather than a domain-specific model

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, llm-inference, nlp, image-processing, rag
- domain: artificial-intelligence, large-language-models, healthcare, computer-vision
- platform: python
- tags: medical-ai, vision-language-model, multimodal, medical-imaging, open-source-model, vllm, healthbench, natural-language-processing, gpu, linux

## Member repositories
- ZJUI-AI4H/Hulu-Med (main) score 62

## Provenance
- Observed fields: from GitHub, fetched 2026-09-03T02:15:08.692171+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:06:39.058312+00:00, confidence not recorded.
  - readme: https://github.com/ZJUI-AI4H/Hulu-Med (fetched 2026-09-03T02:15:08.692171+00:00, sha b4426db89d4b)
- Data as of 2026-08-30T08:39:29.467469+00:00.