Comments
>log in to commentWe built SightRadar to give developers a high‑accuracy face‑recognition API that works as a drop‑in replacement for AWS Rekognition—same request/response shapes, pay‑per‑photo pricing, and GDPR‑ready handling. It lets you index, search, compare and detect faces at scale without changing your code. Would love your thoughts on the API and pricing model 😊
Production-grade face recognition API for indexing, searching, comparing and detecting faces with AWS Rekognition-compatible calls.
- for
- Developers needing scalable face-recognition without rewriting existing Rekognition code.
- pricing
- freemium · free trial
Key features
- Rekognition-compatible API — Same request/response shapes as AWS Rekognition, just change the endpoint URL.
- Index Faces — Detect and store face embeddings in collections for later search.
- Search by Selfie — Find all photos a person appears in from a single selfie.
- 1:1 Compare — Verify identity between two faces with configurable confidence thresholds.
- Detect Only — Locate faces and quality-gate without persisting any data.
- Pay-as-you-go Pricing — Billed per photo processed; no subscription, no minimum usage.
Use cases
- Find a guest across thousands of event photos using one selfie
- Verify KYC sign-ups with real-time 1:1 face comparison
- Batch index large photo collections for searchable galleries
SightRadar pricing
- Real-time$0.00093/photoSynchronous REST response · Always-warm fleet · Zero queueing delay
- Standard (batch)$0.00062/photoUp to 1,000 photos per call · Webhook results · Lower latency cost
- Search by face-id$0.00062/searchVector lookup only · Real-time response
- Storage$6/million faces per monthCharged daily, pro-rated · Delete collection to stop charges
SightRadar vs alternatives
 SightRadar |  Elasticsearch |  Algolia |  Meilisearch |  Replicate | |
|---|---|---|---|---|---|
| Best for | Drop-in Rekognition replacement | Full-text and analytics search | Instant text search | SMS/voice verification | Custom model deployment |
| Pricing | Freemium | Subscription | Subscription | Free | Subscription |
| DevHunt upvotes | 1 | 0 | 0 | 0 | 0 |
| Launched | Oct 2030 | — | — | — | — |
- SightRadar vs Elasticsearch: Elasticsearch provides general search and analytics, not a dedicated face-recognition API.
- SightRadar vs Algolia: Algolia offers text-based search APIs, not biometric face matching.
- SightRadar vs Meilisearch: Vonage verification APIs handle phone-based identity checks, not image-based face verification.
- SightRadar vs Replicate: Replicate runs arbitrary AI models but requires you to host a face model; SightRadar offers a ready-made face API.
SightRadar FAQ
How do I migrate from AWS Rekognition?+
Use the same SDK calls and set the endpoint_url to https://api.sightradar.com; no code changes required.
What is billed for a request?+
One processed photo; a photo with multiple faces costs the same as one with a single face. Compare calls charge for each image.
Are there any minimums or subscriptions?+
No minimum usage, no subscription, and no auto-renewal; credits are prepaid and can be topped up manually.
Is face data stored or used to train models?+
Indexed vectors are stored only in your collection and can be deleted anytime; data is never used to train other models.
What happens if a call fails?+
Failed calls before processing are not charged; you receive a 402 Payment Required if credits run out mid-job.
Does storage incur extra cost?+
Yes, $6 per million stored face vectors per month, charged daily and stopped when the collection is deleted.
Summarized by DevHunt from sightradar.com · Oct 12, 2026. Details may change; check the official site.
