# Face Recognition APIs and Open-Source Libraries in 2026: Prices, Licences and Traps

> Amazon Rekognition, Azure AI Face, Face++, Luxand, DeepFace, InsightFace, CompreFace, plus the FaceCheck.ID and Lenso.ai web search APIs: price per 1,000 searches, free tiers, licences and access limits. Prices and terms from vendor pages and repos, November 12, 2026.

- Source: https://www.stork.ai/blog/face-recognition-api-2026
- Publisher: Stork.AI editorial team
- Published: 2026-11-12
- Updated: 2026-11-12
- Language: en
- Reading time: 7 min

## Key takeaways

- Rekognition: about $1.00 per 1,000 face searches, self-serve.
- Azure Face: similar price, but identify and verify need Microsoft's Limited Access approval.
- InsightFace code is MIT; its pretrained models are non-commercial research only, and that follows them into DeepFace and CompreFace.
- Face++ maker Megvii has been on the US Entity List since October 2019.
- Web face search APIs cost $300 to $560 per 1,000 searches (FaceCheck.ID, Lenso.ai).

**Short answer:** for matching faces against photos **you** supply (log-in, verification, finding a person in your own library), [Amazon Rekognition](https://stork.ai/en/amazon-rekognition) is the cheapest mainstream cloud API at about **$1.00 per 1,000 searches**, and it is self-serve. [Azure AI Face](https://stork.ai/en/azure-ai-face) costs about the same but its identification and verification features need Microsoft's approval. If you want to self-host, [DeepFace](https://stork.ai/en/deepface) (MIT) is the easiest start, but check the model you pick: [InsightFace](https://stork.ai/en/insightface)'s popular pretrained models are licensed for **non-commercial research only**, and that restriction follows them into the libraries that wrap them. Searching the **public web** for a face is a different product at a different price: about $300 per 1,000 searches on [FaceCheck.ID](https://stork.ai/en/facecheck-id) and $560 on [Lenso.ai](https://stork.ai/en/lenso-ai).

**How we checked:** prices come from each vendor's pricing page or public price feed, and licence terms from each repository's own README and licence files, read on November 12, 2026 (US list prices). Costs per 1,000 are our arithmetic, shown in full. We did not benchmark accuracy.

## Two different jobs

- **Gallery matching:** you enrol faces (users, employees, your photo library) and compare new photos against them, 1:1 (is this the same person?) or 1:N (who in my gallery is this?). Rekognition, Azure Face, Face++, Luxand and the open-source libraries do this. It costs fractions of a cent per call.
- **Web face search:** you send a face and get back public web pages where similar faces appear. Only a few services sell this as an API, it costs hundreds of times more per call, and it carries the legal exposure covered in our [guide to face recognition laws](https://stork.ai/blog/face-recognition-laws-2026). Our [face search engine comparison](https://stork.ai/blog/best-face-search-engines-2026) covers the consumer side.

## Price per 1,000 searches

| Provider | Type | Per 1,000 searches or comparisons | Free tier | Catch |
| --- | --- | --- | --- | --- |
| Amazon Rekognition | Cloud API | $1.00 (1,000 x $0.001; falls to $0.0004 at volume) | 12 months: 1,000 images a month per API group | AWS terms restrict police use and sustained surveillance |
| Azure AI Face | Cloud API | About $2.00 (Detect + Identify = 2 transactions at $0.001) | 30,000 transactions a month, but no identify or verify | Identify and verify need Limited Access approval |
| Face++ (Megvii) | Cloud API | $2.00 (1,000 x $0.002, pay as you go) | Free key, shared 3 QPS | Megvii is on the US Entity List; failed calls are billed |
| Luxand.cloud | Cloud API, monthly plans | $0.50 to $6.00, depending on plan, if you use the full quota | Trial; its pages disagree on a free tier | Faces stored capped per plan; no overage price |
| DeepFace | Open source (Python) | $0 plus your hardware; hosted deepface.dev $2.00 | Library free | MIT, but wrapped models keep their own licences |
| InsightFace | Open source | $0 for research; commercial model licence price not published | Code free | Pretrained models non-commercial only |
| CompreFace | Open source (Docker) | $0 plus your hardware | Free | No release since August 2023 |
| FaceCheck.ID API | Web face search | $300 (3 credits x $0.10 per search) | Test mode, no credits, not meaningful results | Credits paid in crypto only |
| Lenso.ai Developer | Web face search | $560 ($2,800 a month for 5,000 requests) | None for the API | Monthly commitment |

*US list prices from each vendor's page or price feed, November 12, 2026. Per-1,000 figures are our arithmetic. Azure's two-transactions-per-search figure is our reading of Microsoft's docs, which say detection is required before other face operations.*

## Cloud APIs

### Amazon Rekognition: cheapest self-serve option

[Amazon Rekognition](https://stork.ai/en/amazon-rekognition) bills per image per API: 1,000 SearchFacesByImage or CompareFaces calls cost $1.00 on the first tier, falling to $0.0004 per image above 35 million a month ([pricing](https://aws.amazon.com/rekognition/pricing/)). Building the gallery costs the same per face indexed, plus $0.00001 per face per month for storage, so 100,000 enrolled faces cost about $1 a month to keep. Count calls, not searches: running DetectFaces and then SearchFacesByImage on one photo is two billed images. The [AWS Service Terms](https://aws.amazon.com/service-terms/) carry a moratorium on police use of face comparison in criminal investigations (missing persons excepted), forbid sustained surveillance of a person without independent review, and add a biometric notice-and-consent addendum.

### Azure AI Face: strong features, gated access

[Azure AI Face](https://stork.ai/en/azure-ai-face) lists $1.00 per 1,000 transactions on its first Standard tier, down to $0.40 at 100 million or more ([pricing](https://azure.microsoft.com/en-us/pricing/details/cognitive-services/face-api/)), plus liveness at $15 per 1,000. Every call is a transaction and Microsoft's docs say detection comes first, so identifying a new photo takes at least two: about $2.00 per 1,000 searches. The bigger issue is access. Identification and verification are [Limited Access](https://learn.microsoft.com/en-us/legal/cognitive-services/computer-vision/limited-access-identity): Microsoft decides at its own discretion, for customers it manages directly and for registered use cases, and not on the free tier. Emotion and gender detection are retired, and age, smile, facial hair, hair and makeup are restricted. Detection alone is open to anyone.

### Face++: cheap, but check your compliance team first

[Face++](https://stork.ai/en/megvii-face) charges $0.002 per Compare or Search call pay-as-you-go, or $1,000 a month for one guaranteed query per second with unlimited calls, which only beats per-call pricing above 500,000 calls a month ([pricing](https://www.faceplusplus.com/v2/pricing-details/)). Failed calls are billed too. Its maker, Megvii, has been on the US Commerce Department's [Entity List since October 9, 2019](https://www.federalregister.gov/documents/2019/10/09/2019-22210/addition-of-certain-entities-to-the-entity-list), which matters for US companies and for many procurement and export-control policies.

### Luxand.cloud: flat monthly plans

[Luxand.cloud](https://stork.ai/en/luxand-cloud) sells monthly plans with request quotas: $9 for 1,500 requests ($6.00 per 1,000), $99 for 200,000 ($0.50 per 1,000) and up ([pricing](https://luxand.cloud/pricing)). Each plan also caps how many faces you can store, from 500 to 54,000, which is the limit most apps hit first. Its own pages disagree on the entry offer (a free 500 requests, a $5 plan and a $9 plan all appear), so confirm in the dashboard.

## Open source

### DeepFace: easiest start, read the model licence

[DeepFace](https://stork.ai/en/deepface) wraps 11 recognition models and 19 face detectors behind a few Python calls (verify, find, represent, analyze), and is MIT-licensed ([repo](https://github.com/serengil/deepface)). Its README warns that the licences of the models and detectors it wraps are inherited, and one of them, Buffalo_L, comes from InsightFace. Pick a model whose own licence allows your use. A hosted version, [deepface.dev](https://deepface.dev/), charges about $2 per 1,000 images.

### InsightFace: MIT code, non-commercial models

[InsightFace](https://stork.ai/en/insightface) is the source of ArcFace, SCRFD and RetinaFace, and its code is MIT. Its [pretrained model packs](https://github.com/deepinsight/insightface#license) (buffalo_l, antelopev2 and others) are for non-commercial research only, whether they download automatically or you fetch them yourself. Commercial use needs a [model licence](https://www.insightface.ai/solutions/face-recognition-licensing) whose price is not published. This is the trap that catches most teams: an MIT badge on the code says nothing about the weights.

### CompreFace: self-hosted, but stalled

[CompreFace](https://stork.ai/en/compreface) gives you a face recognition server with a REST API and a web UI in Docker, under Apache 2.0 ([repo](https://github.com/exadel-inc/CompreFace)). Its default build uses FaceNet; custom builds based on InsightFace carry InsightFace's model terms. Its last release was v1.2.0 on [August 22, 2023](https://github.com/exadel-inc/CompreFace/releases), so treat it as a starting point you will maintain yourself.

## Web face search APIs

Two consumer engines sell API access. [FaceCheck.ID](https://stork.ai/en/facecheck-id)'s [API](https://facecheck.id/Face-Search/API) charges 3 credits per search at $0.10 per credit, about $300 per 1,000 searches, payable in crypto only, with a test mode that searches a small sample and is not meant for real results. [Lenso.ai](https://stork.ai/en/lenso-ai)'s Developer plan is [$2,800 a month for 5,000 requests](https://lenso.ai/en/pricing), about $0.56 each. Both search faces collected from the public web. Building a product on that is legal ground worth reading up on first: the EU AI Act bans building such databases by scraping, and Illinois' BIPA lets individuals sue over face data collected without written consent.

## What to check before you ship

- **Model licence, not just code licence.** InsightFace's weights are non-commercial, and DeepFace and CompreFace can pull them in.
- **Billable calls per user action.** Detect plus search is two calls on Rekognition and Azure; Face++ bills failures.
- **Access approval.** Azure identification and verification need Limited Access; budget weeks, not minutes.
- **Consent and retention.** In Illinois you need written consent and a published retention schedule before creating a face template. AWS and Microsoft both put notice and consent on you in their terms.
- **Vendor status.** Megvii is on the US Entity List; CompreFace is unmaintained.
- **Laws by region.** See our [face recognition laws guide](https://stork.ai/blog/face-recognition-laws-2026) for the EU AI Act, GDPR fines, BIPA and Texas CUBI.

> [!FACT] The price gap is the headline: matching a face against your own gallery costs $1 to $6 per 1,000 searches on the cloud APIs above, while searching the public web for a face costs $300 to $560 per 1,000 through the two APIs that sell it. List prices, November 12, 2026.

## Frequently asked questions

### What is the cheapest face recognition API in 2026?

For matching against faces you supply, Amazon Rekognition at $0.001 per image on its first tier (about $1.00 per 1,000 searches, lower at volume) is the cheapest self-serve option among the major clouds. Luxand's $99 plan works out to $0.50 per 1,000 if you use its whole 200,000-request quota. Self-hosted open source has no licence fee, but check the model licence.

### Can I use InsightFace commercially?

The InsightFace code is MIT-licensed, but its pretrained models such as buffalo_l and antelopev2 are for non-commercial research only. Commercial use needs a separate model licence from InsightFace, and its price is not published.

### Why can't I use Azure Face identification?

Microsoft makes face identification and verification a Limited Access feature: you must register, be a customer managed by Microsoft, and have an approved use case, and it is not available on the free tier. Face detection is open to all Azure customers.

### Is there an API to search the web by face?

Yes, FaceCheck.ID and Lenso.ai sell one. FaceCheck.ID charges 3 credits per search at $0.10 per credit, paid in crypto. Lenso.ai's Developer plan is $2,800 a month for 5,000 requests. Both search faces collected from public websites, which raises legal questions under the EU AI Act and Illinois' BIPA.

### Is DeepFace free for commercial use?

DeepFace's own code is MIT-licensed, but it says the licences of the models and detectors it wraps are inherited. Some are permissive; Buffalo_L comes from InsightFace and is non-commercial. Choose a model whose own licence allows your use.