Specialist modelLimited access
SAM 3
Meta
- Released
- -
- Data date
- October 3, 2026
SAM 3 is Meta’s foundation model for promptable concept segmentation in images and videos. Text, points, boxes, masks, and image exemplars can describe objects. The model detects, segments, and tracks instances instead of generating new images.
The Hugging Face checkpoint is gated and requires consent to share contact information. The weights and official code are intended for local use.
Specifications and access
| Specification | Value and source |
|---|---|
| Model class | Promptable concept segmentation and visual trackingSource |
| Input | Text and visual prompts such as points, boxes, masks, and exemplarsSource |
| Output | Masks, boxes, scores, and object tracks for images and videosSource |
| Access | Gated Hugging Face download after agreement and approvalSource |
| Scope | Segmentation and tracking, not image generationSource |
| License | Other license, details in the Meta model agreementSource |
Measurements without matching peer values
These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.
CountBench accuracy (Show measurement, test conditions, and source)
- Source value
- 93.8
- Score
- 93.8
- Metric
- percent
- Unit
- %
- Category
- object-counting
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- SAM 3 technical paper
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Table 3 object counting.
PixMo-Count accuracy (Show measurement, test conditions, and source)
- Source value
- 86.2
- Score
- 86.2
- Metric
- percent
- Unit
- %
- Category
- object-counting
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- SAM 3 technical paper
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Table 3 object counting.
SA-37 mean IoU, one click (Show measurement, test conditions, and source)
- Source value
- 66.1
- Score
- 66.1
- Metric
- percent
- Unit
- %
- Category
- segmentation
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- SAM 3 technical paper
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Table 6 one-click interactive segmentation.
SA-37 mean IoU, three clicks (Show measurement, test conditions, and source)
- Source value
- 81.3
- Score
- 81.3
- Metric
- percent
- Unit
- %
- Category
- segmentation
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- SAM 3 technical paper
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Table 6 three-click interactive segmentation.
SA-37 mean IoU, five clicks (Show measurement, test conditions, and source)
- Source value
- 85.1
- Score
- 85.1
- Metric
- percent
- Unit
- %
- Category
- segmentation
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- SAM 3 technical paper
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Table 6 five-click interactive segmentation.
SA-Co Gold cgF1 (Show measurement, test conditions, and source)
- Source value
- 54.1
- Score
- 54.1
- Metric
- unknown
- Unit
- No unit provided
- Category
- segmentation
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- SAM 3 technical paper
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Table 1; zero-shot image PCS with text noun-phrase prompt; oracle over three annotations.
- Context
- cgF1 is a source-defined score.
CountBench mean absolute error (Show measurement, test conditions, and source)
- Source value
- 0.12
- Score
- 0.12
- Metric
- mean absolute error (object count)
- Unit
- No unit provided
- Category
- object-counting
- Direction
- Lower is better
- Source type
- vendor-reported
- Evaluator
- SAM 3 technical paper
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Table 3 object counting.
- Context
- The paper expresses this as object-count error. The metric is retained without inventing a standardized unit.
PixMo-Count mean absolute error (Show measurement, test conditions, and source)
- Source value
- 0.21
- Score
- 0.21
- Metric
- mean absolute error (object count)
- Unit
- No unit provided
- Category
- object-counting
- Direction
- Lower is better
- Source type
- vendor-reported
- Evaluator
- SAM 3 technical paper
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Table 3 object counting.
- Context
- The paper expresses this as object-count error. The metric is retained without inventing a standardized unit.
Sources and data date
Every statement links to its underlying documentation or leaderboard.
| Type | Evidence and data date |
|---|---|
| Research status | Research date October 4, 2026. 2 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication. |
| Additional source | Meta SAM 3 model card (retrieved October 3, 2026; October 4, 2026) · Meta SAM 3 model card · Editorial description reviewed October 3, 2026 |
| Additional source | SAM 3 technical paper (retrieved October 4, 2026) |