M
META / sam-3

Segment Anything Model 3

Vision model for detecting, segmenting, and tracking open concepts in images and video through text or visual examples.

01 / SUMMARY

The essential

ORIENTATIONVisual segmentation and tracking

Primary use stated or inferred cautiously from official documentation.

CONTEXTImage or video

Maximum input capacity when published by the source.

EXITMasks and tracking

Documented maximum generation limit.

ACCESSWeights, paper, and playground

Verified availability channels.

MODEL TYPEVision and documents · Video

Primary functional family and verified specialties.

LICENSEMeta model license

Declared terms for API access, model weights, or self-hosting.

02 / TECHNICAL SHEET

Limits and integration

API IDsam-3
Model typeVision and documents · Video
Access modelOpen weights
LicenseMeta model license
DeploymentLocal or private cloud
Release2026
Knowledge cutoffNot published
EntranceText · Image · Video
ExitMasks · Segmented objects
Context windowImage or video
maximum outputMasks and tracking
ReasoningNot applicable or not published
Published tools
Structured OutingsNot applicable or not published
Batch processingNot published
Prompt cacheNot published
Fine-tuningDownloadable weights
Verified platformsSegment Anything Playground · Weight download
BEST SUITED FOR

Visual editing, annotation, video analysis, and concept segmentation.

WORTH MONITORING

It is not an image generator; compare it with segmentation and tracking models.

03 / CAPABILITIES

What it can do

01

Orientation

Visual editing, annotation, video analysis, and concept segmentation.

02

Context

Image or video · Masks and tracking

03

Tools and integration

Not published

04

Access

Weights, paper, and playground

Input modalities
TextImageVideo
04 / PRICES

Documented cost

ConceptWorthUnit/condition
Standard inputNot publishedStandard API rate
Cached inputNot publishedReading reused prefixes
Cache write or storageNot publishedThe condition varies by provider
Standard outputNot publishedMay include reasoning tokens
Batch inputNot publishedAsynchronous processing
Batch outputNot publishedAsynchronous processing

Consult the primary source: the billing unit depends on the model type and access channel.

i Prices change and may depend on level, region or context length. Check the source before making a decision.

05 / EVALUATIONS

How to read the results

“

A public benchmark provides guidance, but does not replace an evaluation with your data, tools, budget, and error tolerance.

BenchmarkResultMetricSource
SA-CoPublished datasetConcept segmentationView source ↗

Inferama only highlights a “best result” when the metric, test set, configuration, and date allow for an equivalent comparison. The supplier's figures are presented as claims from its own source.

06 / VERSIONS

Chronology

Segment Anything Model 3

Version added to Inferama's verified catalog.

07 / SOURCES

Traceability