Model reference · open weights

Hulu-Med

Available as managed deployment LLMs ZJU-AI4H · community Vision + text 1 variants 4k dl/mo

Hulu-Med is an open-weight language model from ZJU-AI4H. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.

Available as managed deployment — configured and operated for you on dedicated EU hardware, quoted per deployment.

What it is

Released byZJU-AI4H
TypeLanguage models
TaskVision + text
Parameters (lead)4.8B
Context256k tokens
Runs withtransformers
Released2025-11-18
Popularity4k downloads / month
LicenceOpen weights

About

What Hulu-Med is

Hulu-Med: A Transparent Generalist Model towards Holistic Medical Vision-Language Understanding

📄 Paper | 🤗 Hulu-Med-4B | 🤗 Hulu-Med-7B |🤗 Hulu-Med-14B |🤗 Hulu-Med-32B | 🔮 ModelScope Models | 📊 Demo

Read the full model card

🔥 News

  • [2025-11-27]Hulu-Med is now compatible with the latest vLLM, offering faster inference and tensor parallel support! Thank you all for your patience and feedback 💪 see here for installation

  • [2025-11-18] 🎊 We released Hulu-Med-4B, a lightweight model with strong multimodal and text reasoning abilities that surpasses MedGemma-4B and Lingshu-7B!

  • [2025-11-01] 📊 Releasing our new evaluation code, MedUniEval! Built on MedEvalKit, MedUniEval is designed for the comprehensive evaluation of medical visual-language models across various modalities—including text, 2D, 3D, and video. More benchmarks are coming soon.

  • [2025-10-15] 🎉 Hulu-Med now supports Transformers integration! HuggingFace-compatible models released with simplified loading and inference. Integration with VLLM is ongoing. The HF models are now available in the main branch on Hugging Face.

  • The model has been updated in the main branch of our Hugging Face repository. You can now load it directly using AutoModelForCausalLM.from_pretrained - the weights will be automatically downloaded.

  • [2025-10-08] Hulu-Med models and inference code released!

📖 Overview

Hulu-Med is a transparent medical vision-language model that unifies understanding across diverse modalities including medical text, 2D/3D images, and videos. Built with a focus on transparency and accessibility, Hulu-Med achieves state-of-the-art performance on 30 medical benchmarks while being trained entirely on public data.

Key Features

  • 🌟 Holistic Multimodal Understanding: Seamlessly processes medical text, 2D images, 3D volumes, and surgical videos
  • 🔓 Fully Transparent: Complete open-source pipeline including data curation, training code, and model weights
  • 📊 State-of-the-Art Performance: Outperforms leading open-source models and competes with proprietary systems
  • Efficient Training: Only 4,000-40,000 GPU hours required for 7B-32B variants
  • 🗂️ Comprehensive Coverage: Trained on 16.7M samples spanning 12 anatomical systems and 14 imaging modalities
  • 🤗 Transformers Native: Now with native HuggingFace Transformers support for easier integration

Comprehensive Data Coverage

Our training corpus encompasses:

  • 12 Major Anatomical Systems: Multi-System, Skin/Integumentary, Respiratory, Cellular/Tissue Level, Digestive, Nervous, Cardiovascular, Musculoskeletal, Reproductive, Urinary, Whole Body, Endocrine, Immune/Lymphatic, and Hematologic systems
  • 14 Medical Imaging Modalities: CT, MRI, X-Ray, Ultrasound, PET, OCT, Endoscopy, Microscopy, Histopathology, Fundus, Dermoscopy, Angiography, Digital Photograph, and Medical Chart
  • Diverse Downstream Tasks: Medical Dialogue, Anomaly Detection, Prognosis Prediction, Treatment Planning, Surgical Skill Assessment, Education, Medical Report Generation, Surgical Phase Recognition, Medical Computation, and more

🏆 Performance Highlights

Medical Multimodal Benchmarks

Performance comparison on medical multimodal benchmarks (For the 'Medical VLM < 10B' subgroup, bold indicates the best method):

ModelsOM.VQAPMC-VQAVQA-RADSLAKEPathVQAMedXQAMMMU-Med
Proprietary Models
GPT-4.175.555.265.072.255.545.275.2
GPT-4o67.549.761.071.255.544.362.8
Claude Sonnet 465.554.467.670.654.243.374.6
Gemini-2.5-Flash71.055.468.575.855.452.876.9
General VLMs < 10B
Qwen2.5VL-7B63.651.963.266.844.120.150.6
InternVL2.5-8B81.351.359.469.042.121.753.5
InternVL3-8B79.153.865.472.848.622.459.2
General VLMs > 10B
InternVL3-14B78.954.166.372.848.023.163.1
Qwen2.5V-32B68.254.571.871.241.925.259.6
InternVL3-38B79.856.665.472.751.025.265.2
Medical VLMs < 10B
LLaVA-Med-7B34.822.746.651.935.220.828.1
MedGemma-4B70.749.272.378.248.125.443.2
HuatuoGPT-V-7B74.353.167.668.144.823.249.8
Lingshu-7B82.956.367.983.161.926.7-
Hulu-Med-4B81.664.671.685.060.126.450.5
Hulu-Med-7B84.266.878.086.865.629.051.4
Medical VLMs > 10B
HealthGPT-14B75.256.465.066.156.724.749.6
HuatuoGPT-V-34B74.056.661.469.544.422.151.8
Lingshu-32B83.457.976.786.765.530.9-
Hulu-Med-14B85.168.976.186.564.430.054.8
Hulu-Med-32B84.669.481.485.767.334.060.4

Medical Text Benchmarks

Performance comparison on medical text benchmarks (bold indicates the best method in each subgroup):

ModelsMMLU-ProMedXQAMedbulletsSGPQAPubMedQAMedMCQAMedQAMMLU-Med
*Proprietary Models

From the published model card. Full card on the HuggingFace links in the sidebar.

Using it via the API

Call it like any OpenAI endpoint

Once AxForge deploys hulu-med for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (hulu-med below is illustrative; you get the exact model name on deployment.)

$ curl -sS https://api.axforge.ai/v1/chat/completions \
  -H "Authorization: Bearer $AXFORGE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"hulu-med","messages":[{"role":"user","content":"Hello"}]}'

Create an account — your API key is available in the console. 3M free tokens every 30 days with every new account.

© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms