Pillar-0 (finetune code + HF weights)

Open 3D radiology foundation model; finetuning framework, weights on Hugging Face

Group

Foundation models & agents

Type

Foundation model

Modality

Radiology: CT (chest, abd-pelvis, head), breast MRI

Maintainer

YalaLab

License

ECL-2.0

Last update

2026-10-12 (last repository push, UTC)

Links

• github.com/YalaLab/pillar-finetune
• huggingface.co/collections/YalaLab/pillar-0

License review: Repo LICENSE is Educational Community License 2.0 (Apache 2.0 with a narrowed patent grant). Hugging Face weights (Pillar0-ChestCT, -AbdomenCT, -HeadCT, -BreastMRI) are tagged ecl-2.0.

Related tools in Foundation models & agents

• 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
• MedRAX: Reasoning agent for chest X-rays that orchestrates multiple CXR models (ICML 2025)

← Back to directory