# Department of Energy labs run Meta's SAM 3 and DINOv3 on 300 A100s to cut beamline image segmentation from a month to about 15 minutes

The SYNAPS-I project, led by Lawrence Berkeley National Laboratory under the Department of Energy's Genesis Mission, fine-tunes Meta's open vision models for detectors that now capture 100,000 images per second.

- Published: 2026-10-05T06:00:43.302Z
- Canonical: https://polylog.news/ai/2026-10-05/department-of-energy-labs-run-meta-s-sam-3-and-dinov3-on-300
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [Meta AI (Genesis Mission)](https://ai.meta.com/blog/genesis-mission-lawrence-berkeley-national-laboratory-segment-anything-dino/), [Meta AI (University of Pittsburgh)](https://ai.meta.com/blog/assistive-robotics-university-of-pittsburgh-sam-dino/), [Berkeley Lab](https://newscenter.lbl.gov/2026/07/22/berkeley-lab-to-lead-13-new-genesis-mission-ai-projects/)

The Department of Energy (DOE) runs light and neutron sources that now produce tens of petabytes of data a year. DOE's laboratories are using Meta's open vision models to process that data. Meta reports that Segment Anything Model 3 (SAM 3)…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-10-05/department-of-energy-labs-run-meta-s-sam-3-and-dinov3-on-300 (subscription information: https://polylog.news/pricing).