> For the complete documentation index, see [llms.txt](https://docs.roboflow.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.roboflow.com/changelog/explore-by-month/june-2026/sam-3-point-prompting-and-video-tracking-in-workflows.md).

# SAM 3 point prompting and video tracking in Workflows

Two new SAM 3 blocks are now available in Workflows. 

The SAM 3 Interactive block adds point and box prompting: pass positive and negative points, or bounding boxes from another model, to segment a specific object and refine the mask. 

The SAM 3 Video Tracker block runs SAM 3 frame by frame on a video stream and keeps per-video memory, so object identities are preserved across frames. You give the video tracker the concepts to follow as text, such as person or forklift, and it picks up matching objects as they enter the scene and assigns each one a tracker ID, with no upstream detector required.

[See the SAM 3 Interactive block documentation](https://inference.roboflow.com/workflows/blocks/sam3_interactive/) 

[See the SAM 3 Video Tracker block documentation](https://inference.roboflow.com/workflows/blocks/sam3_video_tracker/)