Revision History
Segment Anything Model and Dataset (SAM and SA-1B) · 10 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Correction: Corrected SAM 2's speed figure (about 44 FPS for Hiera-B+ video segmentation on an A100, not image segmentation on an H100), MedSAM's training recipe (prompt encoder frozen, image encoder and mask decoder fine-tuned) and dataset size, PerSAM-F (two trained weights, not LoRA), MobileSAM's speedup, the description of SAM's 23-dataset zero-shot evaluation and SA-1B's geographic coverage, an impossible chronology claim about OWL-ViT and OpenSeeD, and consumer features that Meta had only announced (plus 30 minor fixes)
Version 10 · Nov 4, 2026, 08:31 PM
- Improve internal links (2026-10-13)
Version 9 · Oct 13, 2026, 10:06 AM
- Added 6 contextual internal links
Version 8 · Jul 23, 2026, 03:03 PM