What does SightRadar do?

SightRadar is a face recognition API built as a drop-in replacement for AWS Rekognition. It indexes, searches, compares and detects faces, so you can find a person across thousands of photos from one selfie or run 1:1 verification with a similarity threshold you choose.

How much does SightRadar cost?

Billing is per photo, never per face: batch indexing is $0.00062 per photo, real-time is $0.00093 per photo, search by face ID is $0.00062 and face storage is $6 per million faces per month. You start free with trial credits and no card, then pay as you go with prepaid credits, no subscription and no minimums.

How do I migrate from AWS Rekognition?

Point the AWS SDK (boto3 and others) at SightRadar's endpoint. It uses the same request and response shapes, including CreateCollection, IndexFaces, SearchFacesByImage, CompareFaces and DetectFaces, so your existing code keeps working.

How is face data handled?

Face vectors are isolated and encrypted per account and are never used for model training. SightRadar is GDPR-ready and offers a Data Processing Agreement.

What is SightRadar used for?

Consented use cases such as delivering each guest's event or wedding photos from one selfie, selfie-based identity verification for KYC and onboarding, and searching people across your own media library or archive.

PromotedAdvertise

Ask AI

SightRadar

VerifiedWebLive this week

Face recognition API, Rekognition-compatible, pay per photo

face recognition face recognition api face search face verification face detection computer vision aws rekognition alternative rekognition compatible

SightRadar TeamOct 26, 2026Dev toolsDomain control verified

Visit

About

SightRadar is a high-accuracy face recognition API built as a drop-in replacement for AWS Rekognition. It uses the same request and response shapes (CreateCollection, IndexFaces, SearchFacesByImage, CompareFaces, DetectFaces), so you migrate by changing the endpoint URL. Find a person across thousands of photos from one selfie, run 1:1 verification with your own similarity threshold, or detect faces with confidence scores. Billing is per photo, never per face, with no subscription or minimums.

The case

Who it is for

Developers adding face features to their own product: photographers delivering each guest's event photos from one selfie, apps running selfie-based KYC and onboarding, media teams searching people in their own archives, and teams on AWS Rekognition who want a smaller bill without a rewrite.

What is broken today

Face search, verification and detection get expensive once photo volumes grow. Moving to another provider usually means rewriting the integration, retesting every call and rebuilding collections, so most teams stay put and keep paying. Teams handling faces also need that data kept private, isolated per account and out of model training.

What it does about it

SightRadar uses the same request and response shapes as AWS Rekognition (CreateCollection, IndexFaces, SearchFacesByImage, CompareFaces, DetectFaces), so you point the AWS SDK (boto3 and others) at SightRadar's endpoint and your code keeps working. Index photos into a collection, then search it with one selfie to find a person across thousands of photos, compare two faces against a similarity threshold you choose, or detect faces with confidence scores. Photos with no faces return valid empty results, not errors. SDKs are on npm and PyPI, plus a Rekognition shim for Python.

Why not the obvious alternative

The obvious alternative is staying on AWS Rekognition. SightRadar keeps the same API, so moving over is an endpoint change, and it bills per photo, never per face: batch indexing from $0.00062 per photo and real-time at $0.00093 per photo, on prepaid pay-as-you-go credits with no subscription and no minimums. Face vectors are isolated and encrypted per account and never used for model training, and it is GDPR-ready with a Data Processing Agreement. You can start free with trial credits, no card required.

Gallery

Details

CategoryDev tools

PricingFree tier plus paid plans

LaunchedOct 26, 2026

Runs onWeb, API

Built by the makerYes

Evidence

Every claim carries a source and a date

domain: verified
Source badge · verified 2026-10-26

Considered instead of

AWS RekognitionAzure Face APIFace++Google Cloud Vision

Named by the maker, not measured by us. It tells you what this is competing with, not which one wins.

Questions

What does SightRadar do?: SightRadar is a face recognition API built as a drop-in replacement for AWS Rekognition. It indexes, searches, compares and detects faces, so you can find a person across thousands of photos from one selfie or run 1:1 verification with a similarity threshold you choose.
How much does SightRadar cost?: Billing is per photo, never per face: batch indexing is $0.00062 per photo, real-time is $0.00093 per photo, search by face ID is $0.00062 and face storage is $6 per million faces per month. You start free with trial credits and no card, then pay as you go with prepaid credits, no subscription and no minimums.
How do I migrate from AWS Rekognition?: Point the AWS SDK (boto3 and others) at SightRadar's endpoint. It uses the same request and response shapes, including CreateCollection, IndexFaces, SearchFacesByImage, CompareFaces and DetectFaces, so your existing code keeps working.
How is face data handled?: Face vectors are isolated and encrypted per account and are never used for model training. SightRadar is GDPR-ready and offers a Data Processing Agreement.
What is SightRadar used for?: Consented use cases such as delivering each guest's event or wedding photos from one selfie, selfie-based identity verification for KYC and onboarding, and searching people across your own media library or archive.

Discussion

No comments yet. Be the first to give the maker feedback.