Skip to content
SPECIFICTION
Get listed

Face++

faceplusplus.com

Free plan WebWindowsmacOSiOSAndroidAPI
Visit faceplusplus.com
Spec score 8.8 #5 of 37 in AI Image Recognition Software

What it is

Face++ provides computer vision APIs and SDKs for adding image analysis and recognition to applications. Its face capabilities include detection, 1:1 comparison, 1:N search, facial landmarks and facial attributes. The platform also lists body and skeleton recognition, gesture recognition, OCR, image beautification and photo album clustering. Liveness detection distinguishes a live face from a photo, video or mask to help prevent spoofing attacks. Face++ presents identity verification, eKYC and passport matching, ride-hailing driver verification, biometric retail payments and skin analysis as use cases. Its platforms include Android, API, iOS, macOS, web and Windows. The Face Landmark SDK supports Android, iOS, Windows and Mac, and can perform face detection, tracking and landmarking locally on mobile devices. The free API key has shared QPS limits: 3 total for facial recognition APIs, with Dense Facial Landmarks API limited separately to 1. The SDK trial offers one day of authorized online licensing, with up to five authorizations per day. The free plan is free; other listed billing options have no listed price.

Spec sheet

What it does
Face++ is an AI computer-vision platform offering facial, body, image, object and text-recognition APIs and SDKs.Read 21 Oct 2026 · faceplusplus.com
Support
Face++ offers technical support for integration help, API issues, and troubleshooting through its contact page.Read 8 Oct 2026 · faceplusplus.com
Headquarters
Haidian District, Beijing, ChinaRead 28 Sep 2026 · faceplusplus.com
Founded
2012Read 28 Sep 2026 · faceplusplus.com
API access
Developers access web APIs through Face++ API keys and secrets, with documented POST endpoints and JSON responses.Read 21 Oct 2026 · console.faceplusplus.com
Body recognition
The platform lists body detection, skeleton detection, body outlining, body attributes and gesture recognition.Read 21 Oct 2026 · faceplusplus.com
Company founded
Megvii was incorporated in 2011 by Tang Wenbin, Yang Mu and Yin Qi.Read 21 Oct 2026 · crunchbase.com
Company headquarters
Crunchbase lists Megvii in Beijing, China, while the World Economic Forum places its headquarters in Zhongguancun.Read 21 Oct 2026 · crunchbase.com
Company history
Face++ says its first version launched in 2012 and the Face++ AI Open Platform was announced in 2017.Read 8 Oct 2026 · faceplusplus.com
Company legal name
Crunchbase identifies the maker as Megvii Technology Limited, also known as 旷视 and Face++.Read 21 Oct 2026 · crunchbase.com
Company size
Crunchbase places Megvii in the 1,001–5,000 employee range.Read 21 Oct 2026 · crunchbase.com
Data residency
For clients outside China, personal information may be processed on servers in China or Canada according to the client’s choice.Read 21 Oct 2026 · faceplusplus.com
Documentation
Face++ provides a console dashboard, documentation center, API documents, FAQ and usage and billing views.Read 21 Oct 2026 · faceplusplus.com
Face attributes
Face attributes include age, gender, smile, head pose, eye status, emotion, beauty, gaze, mouth status, skin status and image quality.Read 21 Oct 2026 · faceplusplus.com
Face detection
The Detect API locates faces in images and returns bounding boxes and face tokens.Read 21 Oct 2026 · faceplusplus.com
Face recognition
Its capabilities include face detection, 1:1 comparison, 1:N search, facial landmarks, and facial attributes.Read 8 Oct 2026 · faceplusplus.com
Face search
Its 1:N search finds similar faces in a collection and returns similarity confidence information.Read 21 Oct 2026 · faceplusplus.com
Face verification
Its 1:1 verification compares two faces to determine whether they belong to the same person.Read 21 Oct 2026 · faceplusplus.com
Facial landmarks
Face++ provides ordinary and dense facial-landmark detection, including 106-point landmarks in its Landmark SDK.Read 21 Oct 2026 · faceplusplus.com
Free API limits
The free API key has shared QPS limits, including 3 total for facial recognition APIs, with Dense Facial Landmarks API limited separately to 1.Read 8 Oct 2026 · faceplusplus.com
Funding news
Crunchbase News reported on May 8, 2019 that Megvii announced a $750 million financing round and had raised $1.4 billion since founding.Read 21 Oct 2026 · news.crunchbase.com
Global reach
The World Economic Forum states that Face++ and FaceID have users in 220 countries and regions.Read 21 Oct 2026 · weforum.org
Identity solutions
Listed identity use cases include KYC, KYB, anti-spoofing, liveness detection, deepfake detection and AML-risk data.Read 21 Oct 2026 · faceplusplus.com
Image beautification
Face merging, filters and AR-sticker SDKs support image beautification and face effects.Read 21 Oct 2026 · faceplusplus.com
Industry integrations
The maker describes applications in fintech, public security, government, smart campuses and offices, banking, lending, delivery, automotive and education.Read 21 Oct 2026 · faceplusplus.com
Integration
The Face Landmark SDK can run face detection, tracking, and landmarking locally on mobile devices.Read 8 Oct 2026 · faceplusplus.com
Licence
The developer agreement grants a revocable, non-transferable, non-sublicensable, non-exclusive service licence and reserves other rights to Megvii.Read 21 Oct 2026 · faceplusplus.com
Limit: accuracy variability
A G2 reviewer reported that the system sometimes failed to recognize a face despite perceived accuracy around 99 percent.Read 21 Oct 2026 · g2.com
Limit: API quotas
The January 2026 overseas policy caps monthly free calls, including 500 detect calls, 500 compare calls and 500 search calls.Read 21 Oct 2026 · faceplusplus.com
Limit: documentation depth
A G2 reviewer asked for more support and help documents, while another found the many detection categories difficult to choose among.Read 21 Oct 2026 · g2.com
Limit: image size
Documented APIs reject image files larger than 2 MB with a request-entity-too-large or image-file-too-large error.Read 21 Oct 2026 · console.faceplusplus.com
Limit: setup complexity
A G2 reviewer said setup appeared complicated and could be error-prone, including when makeup was involved.Read 21 Oct 2026 · g2.com
Liveness
Liveness detection is described as distinguishing a live face from a photo, video, or mask to help prevent spoofing attacks.Read 8 Oct 2026 · faceplusplus.com
Offline processing
Face Compare SDK supports offline face detection, comparison, tracking and landmarks.Read 21 Oct 2026 · faceplusplus.com
Other vision features
The platform also lists body and skeleton recognition, gesture recognition, OCR, image beautification, and photo album clustering.Read 8 Oct 2026 · faceplusplus.com
Platform
Face++ offers computer vision APIs and SDKs for adding image analysis and recognition to applications.Read 8 Oct 2026 · faceplusplus.com
Policy date
The Face++ privacy policy states it was last updated on September 30, 2020.Read 8 Oct 2026 · faceplusplus.com
Praise: documentation
A G2 reviewer described the documentation as helpful and setup as easy.Read 21 Oct 2026 · g2.com
Praise: ease and speed
G2 reviewers repeatedly praised intuitive use, easy setup, fast processing and accurate face bounding boxes.Read 21 Oct 2026 · g2.com
Rating on G2
3.7 out of 5 from 6 reviewsRead 21 Oct 2026 · g2.com
Recent maintenance notice
Face++ announced scheduled infrastructure upgrade and service migration work for September 15, 2026.Read 21 Oct 2026 · faceplusplus.com
SDK availability
SDK products include Face Landmarks, Dense Facial Landmarks, Face Compare, Skeleton Detection, Portrait Segmentation, Filter and AR Sticker SDKs.Read 21 Oct 2026 · faceplusplus.com
SDK performance
The Landmark SDK page reports 100 ms detection and 7 ms tracking at 1080p, with a roughly 5 MB package.Read 21 Oct 2026 · faceplusplus.com
SDK platforms
The Face Landmark SDK lists Android, iOS, Windows and Mac support.Read 21 Oct 2026 · faceplusplus.com
SDK trial
The SDK trial provides one day of authorized online licensing for Landmark SDK on Android and iOS, with up to five authorizations per day.Read 8 Oct 2026 · faceplusplus.com
Security and privacy
The privacy policy says Face++ processes facial and body biometric information and supports access, correction and deletion rights.Read 21 Oct 2026 · faceplusplus.com
Self-hosting
The public terms describe hosted web APIs and SDK services and do not grant a self-hosted deployment licence.Read 21 Oct 2026 · faceplusplus.com
Text recognition
Face++ offers universal OCR, template OCR and recognition for identity cards, driving licences, vehicle licences and bank cards.Read 21 Oct 2026 · faceplusplus.com
Use cases
The site presents identity verification, eKYC and passport matching, ride-hailing driver verification, biometric retail payments, and skin analysis as use cases.Read 8 Oct 2026 · faceplusplus.com

Plans and prices

Free Free No credit card, upfront fee, or commitment Free API access · shared QPS capacity · one free API key Read 21 Oct 2026 · faceplusplus.com
Free Free Free API access; one API key; free-key QPS limits vary by API group Read 8 Oct 2026 · faceplusplus.com
Daily/Monthly Plan Priced on request Prepay for future usage by QPS capacity Capacity measured in queries or requests per second Read 8 Oct 2026 · faceplusplus.com
Daily/Monthly Plan Price not published Prepaid by QPS capacity Daily or monthly duration · configurable start time · QPS capacity Read 21 Oct 2026 · faceplusplus.com
Pay As You Go Priced on request Top up account and pay for consumption Usage-based billing · API services Read 21 Oct 2026 · faceplusplus.com
SDK licensing Priced on request Billed by license period or license number Face Landmark SDK and Face Compare SDK licensing options Read 8 Oct 2026 · faceplusplus.com
SDK licensing Priced on request Billed by license period or license number Online or offline licensing depending on SDK · contact sales for offline trial Read 21 Oct 2026 · faceplusplus.com

Every Face++ plan and price →

Strong points and limits

Strong points

  • Face APIs cover detection, comparison, search and attributes.
  • Lists OCR, gesture and body recognition.
  • Landmark SDK can run locally on mobile devices.
  • Free API plan and SDK trial are available.

Limits

  • Free API key has shared QPS limits.
  • Landmark SDK trial is limited to one day.
  • Other listed billing options have no listed price.

Who it suits

Face++ suits developers adding image analysis, face recognition or other vision capabilities to applications. Its listed use cases include identity checks, driver verification, retail payments and skin analysis.

Face++ in depth

Read the full write-up

Face++ is a computer-vision platform offering APIs and SDKs for adding face and image analysis to applications. It is a strong fit for developers building identity workflows or combining several vision capabilities; the main caution is that free API throughput is shared and the SDK trial is short.

Overview

Face++ brings face analysis together with body and skeleton recognition, gesture recognition, OCR, image beautification and photo album clustering. That breadth is useful when an application needs more than face matching, though teams seeking a single narrow capability may not need the full platform.

Its stated use cases range from identity verification, eKYC and passport matching to ride-hailing driver checks, biometric retail payments and skin analysis. Liveness detection is designed to distinguish a live face from a photo, video or mask, adding a check against spoofing. These capabilities can support such workflows, but the appropriate performance and compliance fit will depend on the application.

Key features

Face analysis and liveness

The face APIs cover detection, 1:1 comparison, 1:N search, facial landmarks and facial attributes. This mix can support both checking a claimed identity and searching a collection for a matching face. Liveness detection adds a separate anti-spoofing step, which is relevant when a workflow must do more than compare still images.

Broader vision tools and local SDK processing

OCR, body and skeleton recognition, gesture recognition, image beautification and album clustering extend the platform beyond biometrics. The Face Landmark SDK can perform face detection, tracking and landmarking locally on mobile devices, giving teams an on-device option for those tasks. Its SDK lineup includes Face Landmark and Face Compare licensing options.

Integration support and privacy

Face++ offers technical support for integration, API issues and troubleshooting through its contact page. Services process personal information under MEGVII’s privacy policy, which describes collection, use, retention, disclosure, protection and global transfers. The policy was last updated on September 30, 2020, so organizations assessing current privacy obligations should review the policy alongside their own requirements.

Pricing

The Free plan costs free and provides free API access with one API key. Free-key QPS limits vary by API group: facial recognition APIs share a total limit of 3 QPS, while Dense Facial Landmarks has a separate limit of 1 QPS. Those caps suit initial integration and low-throughput work better than a service that must handle sustained volume.

For higher or more predictable throughput, the Daily/Monthly Plan uses prepayment for future usage by QPS capacity, measured in queries or requests per second; it has custom pricing. Pay As You Go requires topping up an account and charges for consumption, also with custom pricing. The choice is between reserving capacity and paying by usage, but teams need a quote to compare either path against their workload.

SDK licensing has custom pricing, billed by license period or license number, with Face Landmark SDK and Face Compare SDK options. The SDK trial gives one day of authorized online licensing for the Landmark SDK on Android and iOS, with up to five authorizations per day. That is enough for a brief evaluation, but a constrained basis for extended testing or repeated authorization.

Platforms

The platform lists Android, iOS, macOS, web and Windows alongside API access. The Face Landmark SDK specifically supports Android, iOS, Windows and Mac. Its local face detection, tracking and landmarking make it relevant to mobile applications that need these operations on-device; the trial licensing described for the SDK is limited to Android and iOS.

Who it's for

Face++ suits developers who need several vision APIs in one platform, particularly teams building identity verification, document matching or face-search workflows. It is also worth considering when mobile face landmarking or additional tools such as OCR and gesture recognition belong in the same application. Teams with low-volume needs can start on the free API key; higher-throughput projects should compare custom-priced capacity or usage billing with their expected demand. It is a weaker fit when the deciding requirement is a long SDK evaluation, since the Landmark SDK trial lasts one day.

Pros and cons

Pro: A broad set of face, body, gesture, OCR and image features can serve applications that would otherwise need several different vision capabilities.

Pro: The Landmark SDK supports local detection, tracking and landmarking on mobile devices, an option for teams that want those operations on-device.

Pro: Free API access and three paid billing approaches—capacity-based, usage-based and SDK licensing—give teams different ways to begin or scale.

Con: Free facial-recognition throughput is shared at 3 QPS, with Dense Facial Landmarks separately capped at 1 QPS, limiting the free key for sustained traffic.

Con: The one-day Landmark SDK trial, capped at five authorizations per day, leaves little room for a prolonged evaluation.

Con: Daily/Monthly, Pay As You Go and SDK licensing all have custom pricing, so costs cannot be compared from a published figure alone.

Alternatives

AI Image Recognition Software is a broader category directory for readers who want to compare options beyond Face++.

Sighthound AI is worth comparing when its paid offering and broad platform coverage, including Linux and self-hosted deployment, better match the target environment; it also has a free plan and trial.

MediaPipe is a better fit for teams seeking a free, on-device ML framework across platforms including Linux and self-hosted deployments, rather than Face++’s API-and-SDK offering.

Google Cloud Vision API is an alternative for image-oriented API workloads with per-image billing and monthly usage tiers; Face++ is more directly positioned around face analysis and related SDKs.

SentiSight.ai merits consideration when an offline model license and self-hosted deployment are priorities; Face++ instead offers APIs and SDKs with custom-priced licensing and usage options.

CompreFace is the clearer choice for a free, open-source, Docker-based deployment with CPU and GPU support, especially when self-hosting is preferred.

InsightFace may suit research and non-commercial projects looking for an open-source option; commercial model use requires a separate license.

Roboflow is an alternative for teams whose stated needs align with its training and inference credits, rather than Face++’s face-focused recognition and landmark SDK options.

Yandex Vision OCR is the more focused option when the requirement is printed text or license-plate recognition; its pricing is paid rather than freemium.

Verdict

Choose Face++ when an application needs a combination of face recognition, liveness and other vision capabilities, with SDK support for mobile and desktop development. Its breadth and local Landmark SDK processing are the clearest reasons to choose it. Look elsewhere if predictable public pricing, higher free throughput or a longer SDK trial is central to the decision.

Verdict

Face++ combines a broad set of vision APIs with SDKs for mobile and desktop platforms. Check the shared free-key limits and the unpriced billing options against the expected workload.

Where Specifiction places it

Alternatives to Face++All alternatives →