# Voio

> Voio is a Berkeley-based frontier AI lab building a unified reading platform for radiology. Spun out of UC Berkeley and UCSF, it develops the open-source Pillar family of models that interpret CT, MRI and X-ray scans across hundreds of conditions, aiming to help radiologists work faster without trading away accuracy. The company exited stealth in November 2025 with $8.6M in seed funding.

- **Founded:** 2025
- **Headquarters:** Berkeley, California, United States
- **Founders:** Adam Yala (Co-founder & CEO; Assistant Professor of Computational Precision Health, UC Berkeley & UCSF), Maggie Chung (Co-founder & Medical Lead; Assistant Professor of Radiology & Biomedical Imaging, UCSF; practicing radiologist), Trevor Darrell (Co-founder; Professor of Computer Science, UC Berkeley; founder of Berkeley AI Research (BAIR))
- **Team size:** ~15 employees
- **Products:** Pillar-0, Unified reading platform, Pillar model family
- **Notable:** Exited stealth in November 2025 with $8.6M seed funding., Released Pillar-0, described as the most accurate publicly available radiology AI model to date., Reported 10-17% accuracy gains over leading models from Google, Microsoft and Alibaba.

## Products & services

- **Pillar-0** — Open-source foundation model that directly interprets CT and MRI scans, recognizing hundreds of conditions. Reported .87 AUC across 350+ findings in chest CT, abdomen CT, brain CT and breast MRI, 10-17% higher accuracy than Google MedGemma (.76), Microsoft MI2 (.75) and Alibaba Lingshu (.70).
- **Unified reading platform** — A single workspace that removes context-switching between clinical systems, letting AI draft comprehensive reports for radiologist review across scan types and tasks.
- **Pillar model family** — Successor models extending coverage across CT, MRI and X-ray and hundreds of pathologies, built for independent validation and clinical integration.

## Achievements

- Exited stealth in November 2025 with $8.6M seed funding.
- Released Pillar-0, described as the most accurate publicly available radiology AI model to date.
- Reported 10-17% accuracy gains over leading models from Google, Microsoft and Alibaba.
- Founding team previously built Mirai (breast cancer risk, validated across 2M+ mammograms) and Sybil (lung cancer risk from screening scans).

## Latest updates

- **2025-11** — Voio emerged from stealth with $8.6M seed funding and launched the open-source Pillar-0 model.
- **2026-04** — UC Berkeley and Berkeley News features on the team using AI to revolutionize medical imaging.

## Links

- Website: https://voio.com
- LinkedIn: https://www.linkedin.com/company/voio-inc
- GitHub: https://yalalab.github.io/pillar-0/

---

Profile page: https://yespress.io/voio
Published by YesPress — https://yespress.io
Last updated: 2026-07-02