# Berkeley Lab Cuts X-Ray Image Segmentation From Weeks to Roughly 15 Minutes Using Meta's Open Vision Models

SAM 3 and DINOv3, fine-tuned on scientific imaging and served across 300 A100 accelerators, now return fully segmented three-dimensional volumes to researchers at the beamline.

- Published: 2026-08-16T06:12:17.236Z
- Canonical: https://polylog.news/ai/2026-08-16/berkeley-lab-cuts-x-ray-image-segmentation-from-weeks-to-rou
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [Meta AI](https://ai.meta.com/blog/genesis-mission-lawrence-berkeley-national-laboratory-segment-anything-dino/), [Berkeley Lab](https://newscenter.lbl.gov/2026/07/22/berkeley-lab-to-lead-13-new-genesis-mission-ai-projects/), [Tech Times](https://www.techtimes.com/articles/321456/20260724/meta-open-source-ai-cuts-doe-beamline-analysis-month-minutes.htm)

A United States Department of Energy project called SYNAPS-I is running Meta's open-weight SAM 3 and DINOv3 models to segment X-ray and neutron imaging data in near real time. Lawrence Berkeley National Laboratory leads the effort with Argo…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-08-16/berkeley-lab-cuts-x-ray-image-segmentation-from-weeks-to-rou (subscription information: https://polylog.news/pricing).