# Berkeley Lab cuts beamline image analysis from a month to minutes with open vision models

The SYNAPS-I project pairs DINOv3 features with Segment Anything masks across 300 A100 GPUs to handle the tens of petabytes that Department of Energy light and neutron sources produce each year.

- Published: 2026-09-15T06:27:29.669Z
- Canonical: https://polylog.news/ai/2026-09-15/berkeley-lab-cuts-beamline-image-analysis-from-a-month-to-mi
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [Meta AI](https://ai.meta.com/blog/genesis-mission-lawrence-berkeley-national-laboratory-segment-anything-dino/)

Lawrence Berkeley National Laboratory's SYNAPS-I project is using Meta's open-source SAM 3 and DINOv3 models against a data problem that has limited experimental science for years, Meta says. Department of Energy light and neutron sources n…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-09-15/berkeley-lab-cuts-beamline-image-analysis-from-a-month-to-mi (subscription information: https://polylog.news/pricing).