SA-V
Video segmentation, cgF1
domainactive
Definition
- Version
- not in the notes
- Category
- domain
- Maintainer
- Meta
- Status
- active
- Home page
- not in the notes
Sources
- GitHub - facebookresearch/sam3: The repository provides code for running inference and finetuning with the Meta Segment Anything Model 3 (SA github.com https://github.com/facebookresearch/sam3
Scores
| Model | Score | Who ran it | Setting | View or population | Date | Source |
|---|---|---|---|---|---|---|
| SAM 3 | 30.3 cgF1 | vendor-reported | test; human 53.1 | – | Nov 2025 | github.com |
Where this comes from
Scores from different settings, harnesses or dates are not directly comparable; the row says which. Sorted by score, highest first, as a reading aid, not a ranking.
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "SA-V", https://aurion.semai.id/benchmarks/sa-v/, records of October 2026.