AI Model Radar
中文 EN

Segment Anything Model 3

Meta
Model overview

SAM 3

Meta's current public SAM 3 open-weight promptable segmentation model for images and video.

segmentationobject trackingvisual segmentationevaluationlocal inference
CompareCompanyRankingsPublic signals
Representative version

SAM 3

CompanyMeta
ReleaseNot disclosed
Release typeModel release
Parameters0.86B
ArchitectureNot disclosed
ContextNot disclosed
Input modalitiesimage / video / text
Output modalitiesstructured_output
Tool use / function callingNot disclosed
Structured output / reasoningConfirmed: Structured output
Vision / audio / videoConfirmed: Vision, Video; Not shown: Audio
Open weights / LicenseWeights available / license pending verification
meta-sam-3-current · License source pending verification
Vendor APINot disclosed
API source pending verification
OpenRouterNot separately listed on OpenRouter
PricingNo reliable public source
BenchmarkNo reliable public source
Public signals1 activity snapshot
Use casesvisual segmentation / research evaluation / local inference / segmentation
Claim evidenceRelease eventfacebook/sam3 Hugging Face model cardSAM 3
Model specs / weightsSegment Anything Model 3 and Segment Anything Model 3.1 are here | Meta AISegment Anything Model 3 · SAM 3
OpenRouter

Pricing / context / availability

OpenRouterNot separately listed on OpenRouter
OpenRouter listingNot separately listed in the current OpenRouter public list
Mapped versionSAM 3
meta-sam-3-current
PricingNo reliable public source
ContextNot disclosed
Speed/LatencyNo reliable public source
Snapshot time2026-07-15T20:11:41.551Z
SourceView source
Public activity counters

Cumulative Hugging Face model-card downloads / likes / spaces

PlatformEntryDownloadsLikesSpacesTimeSource
huggingfacefacebook/sam32,182,2162,4501002026-07-15T20:11:41.551ZView source
Evidence gaps

Public display boundary

Release date not disclosedOpenRouter not separately verifiedNo reliable benchmark sourceNo linkable real reviewSpeed/latency not disclosed