Face++
Image recognition software
Deep learning software
AI face recognition tools
- Features
- Ease of use
- Ease of management
- Quality of support
- Affordability
- Market presence
Take the quiz to check if Face++ and its alternatives fit your requirements.
Pay-as-you-go
Free TrialFree version
Small
Medium
Large
-
What is Face++
Face++ is a cloud-based computer vision API suite focused on face detection, face comparison/verification, face search, and facial attribute analysis. It is used by developers and product teams to embed face-related capabilities into mobile apps, web services, and enterprise workflows via REST APIs and SDKs. The platform also offers broader image analysis features, but its core differentiation is packaged face recognition functions and related identity/attribute endpoints.
Mature face API coverage
Face++ provides a broad set of face-focused endpoints such as detection, landmarking, verification, and 1:N search. This reduces the need to assemble separate models and pipelines for common face recognition workflows. For teams that want an API-first approach, it can be integrated without building a full training and deployment stack.
Developer-friendly integration model
The product is delivered primarily as APIs and SDKs, which fits application teams that need to embed face recognition into existing products. This approach typically shortens implementation time compared with building custom models and hosting infrastructure. It also supports consistent behavior across environments because inference runs on the vendor-managed service.
Broader vision capabilities available
Beyond face recognition, Face++ includes additional computer vision functions (for example, general image analysis features) that can support adjacent use cases. This can be useful when a project needs both face-specific and general image understanding under one vendor account. It may reduce vendor sprawl for teams that would otherwise use separate services for different vision tasks.
Limited model transparency/control
As a managed API, Face++ provides less control over model architecture, training data, and tuning than platforms designed for custom model development. Organizations with strict requirements for explainability, reproducibility, or bespoke performance targets may find the black-box nature constraining. Customization options are typically narrower than end-to-end model training and MLOps toolchains.
Face recognition governance burden
Face recognition use cases often require heightened privacy, consent management, and regulatory review. Buyers may need to implement additional controls around data retention, access logging, and lawful basis for processing. These governance requirements can add time and cost beyond the technical integration.
Potential vendor lock-in risk
Applications built tightly around proprietary face endpoints and response formats can be harder to migrate later. This is especially relevant when workflows depend on vendor-specific face templates, similarity scoring, or search behavior. Teams may need an abstraction layer to reduce switching costs.
Plan & Pricing
Pricing model: Mixed — Permanent free tier + Pay-as-you-go + QPS (time-window) packages + SDK licensing.
Free tier: ¥0 — Register and use Face++ Web API for free (no credit card). Free tier: call all APIs, limited/shared QPS, limited face storage (up to 1000 FaceSet, each FaceSet up to 10,000 face_tokens), single API Key..
Pay-as-you-go (按量计费): Top up account and pay per API call. Example per-call prices (CNY):
- Face detection API: ¥0.001 per call.
- Face analysis API: ¥0.001 per call.
- Face compare (face pair/比对) API: ¥0.01 per call.
- Face search API: ¥0.01 per call.
- Face library management / face data management API: ¥0.0005 per call. (These per-API unit prices and QPS minimums are listed in the official API price table.)
QPS / time-window packages (包时计费): Prepay for QPS capacity for a specified time window (day / month). Example prices (CNY):
- Face-recognition API package: 1 QPS * 1 day = ¥100; 1 QPS * 1 month (30 days) = ¥1,000.
- Typical image/person APIs (examples): 1 QPS * 1 day = ¥100–¥1,000 depending on API; monthly packages shown in the API price table (see SDK/API price list).
Bulk / tiered usage packs (次数包计费): One-year-validity count bundles with volume discounts (examples):
- Face detection: 50 million calls pack at ¥0.001 per call (total ¥50,000 for 50M).
- Face compare/search: tiered packs (250k / 750k / 1.2M / 1.8M) with per-call prices from ¥0.008 down to ¥0.006 depending on volume.
SDK licensing (移动端 SDK 授权):
- Networked (联网) authorization: starting at ¥20,000 per year for 1,000 authorizations per year per platform.
- Offline authorization (离线授权): starting at ¥300,000 per year per platform (unlimited authorizations). -Mobile SDKs: free trial available; offline test licenses available via sales.
Discounts / notes:
- Volume discounts available via count-packs and larger QPS packages; the vendor states “the more you top up / buy, the more bonus/discount you get.”
- Failed calls typically do not incur charge in package modes; pay-as-you-go charges include successful calls and some partial failures — see official price table for details.
(All items above are taken from the official Face++ pricing and API price table pages on faceplusplus.com.cn / faceplusplus.com.)
Seller details
Megvii Technology Limited
Beijing, China
2011
Private
https://www.faceplusplus.com/
https://www.linkedin.com/company/megvii
