# Open-Weight Vision Models Move Into National Lab Beamlines and Assistive Robotics

A Berkeley Lab-led project using Meta's Segment Anything 3 and DINOv3 cut per-time-step image annotation from about a month of expert work to roughly 15 minutes on 300 A100 GPUs.

- Conviction: 49 / 100 (weakening)
- 7-day move: +9
- Horizon: Short term (next 30 days)
- Tracking since: 2026-10-03T00:00:00.000Z
- Last updated: 2026-10-12T14:07:13.413Z
- Canonical: https://polylog.news/ai/trends/event-open-weight-vision-models-move-into-national-lab-beamlines-a
- Publisher: Polylog
- Affected regions: United States

## Recent score history

- 2026-10-11: 51
- 2026-10-12: 49

## Recent evidence

- [confirms] Berkeley Lab Puts Meta's SAM 3 and DINOv3 at the Center of a Genesis Mission Imaging Pipeline, Cutting Analysis From a Month to 15 Minutes (2026-10-11): The Berkeley Lab-led SYNAPS-I project runs fine-tuned SAM 3 and DINOv3 on 300 A100 GPUs across national lab sites under the Genesis Mission, cutting analysis from a month to 15 minutes. It starts with drought tracking in grapevine micro-CT scans, which extends the use of open vision models in national lab science.
- [confirms] Pitt's ARPA-H Robotic Wheelchair Runs Meta's DINO and SAM On-Device for Real-Time 360-Degree Perception (2026-10-11): Pitt's ARPA-H wheelchair pairs a robotic arm with Meta's open DINO and SAM models running on the device. This is the assistive-robotics side of the thesis, in a federally funded program.

6 more evidence entries, the full score history, the conviction-driver timeline, and affected assets are for subscribers: https://polylog.news/pricing