Azure AI Face
Microsoft's face detection and verification API — now gated behind limited-access enrollment, with emotion and gender inference retired.
About
Azure AI Face detects faces, extracts landmarks and quality attributes (blur, exposure, glasses, head pose, occlusion), and offers Verify and Identify operations against person groups. It was once a full attribute-analysis API, but Microsoft's 2022 Responsible AI overhaul retired emotion and gender inference entirely (access revoked June 2023) and restricted age, smile, facial hair, hair, and makeup to approved accessibility scenarios.
Since then the service is limited-access: new customers must apply through a Face Recognition intake form and justify an acceptable use case, and Microsoft bars use by or for US police departments. That makes it the most policy-constrained of the big three cloud face APIs — fine for legitimate verification workloads, but not something you can just sign up and call.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Microsoft publishes no accuracy benchmarks. It does publish fairness guidance (the open-source Fairlearn package and a face-verification fairness notebook) — transparency material, not accuracy claims. No independent published benchmark found.
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 7/10
Scored 7/10 (verified, metrics-v1). Evidence: official Microsoft Learn documentation current as of 2026; the 2022–2023 Responsible AI retirement of emotion/gender attributes is documented in Microsoft's own transparency notes. Strengths: strong responsible-AI documentation, fairness tooling, enterprise-grade verification. Docked for: (1) limited-access enrollment makes it unavailable to casual users; (2) no published accuracy benchmarks and the feature set has shrunk. License: commercial service; no weights available.
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 8.0/10
- Maintenance · recency, releases, issue load
- 8.0/10
- Openness · license + docs
- 3.0/10
- Manual review · sanity read + red-flag check
- 8.0/10
- · Microsoft Learn docs current; limited-access enrollment via Face Recognition intake form since 2022.
- · Emotion and gender inference retired June 2023; age/smile/facial-hair/hair/makeup restricted to approved accessibility scenarios.
- · Microsoft publishes fairness tooling (Fairlearn notebook), not accuracy benchmarks.