Hounsfield LabRadKev
Open-weight radiology decision models (9B and 27B adapters) that score fixed-choice radiology questions in one forward pass without generating text
LICApache-2.0 (code); CC BY-NC-SA 4.0 (weights)UPD2026-10-07STATUSMaintainedCOMMNon-commercial
- Group
- Foundation models & agents
- Type
- Foundation model
- Modality
- Radiology (report and case decision tasks)
- Maintainer
- U. Wisconsin–Madison (Ram / Zhang)
- License
- Apache-2.0 (code); CC BY-NC-SA 4.0 (weights)
- Last update
- 2026-10-07 (last repository push, UTC)
Links
License review: Code is Apache-2.0. Hugging Face adapters (ramu9703/radkev-27b, radkev-9b) are CC BY-NC-SA 4.0 for non-commercial research; gated access requires accepting those terms. Not a medical device.
Commercial use: read the project’s license (and any model-weight terms) yourself before relying on it. This isn’t legal advice.
Related tools in Foundation models & agents
- Pillar-0 (finetune code + HF weights): Open 3D radiology foundation model; finetuning framework, weights on Hugging Face
- Pillar-0 pretraining: Vision-language pretraining code for Pillar-0
- RATE: Uses an LLM (Qwen3 via sglang) to extract structured labels for 366 findings from radiology reports
- Merlin: 3D vision-language model for CT that uses structured EHR and report supervision
- MedGemma: Google's open medical model family; repo has notebooks and code
- CheXagent: Foundation model for chest X-ray interpretation