CompreFace
Free, open-source, self-hosted face recognition server: one docker command gives you a full REST API for detection, recognition, age, and gender.
About
CompreFace is a docker-based face recognition service built on top of the state-of-the-art open libraries (FaceNet, InsightFace families) with a plain REST API and a management UI. It covers face detection, recognition, verification, landmarks, age, gender, mask detection, and head pose, with role-based access control and multi-model support on CPU or GPU.
The whole point is that you need zero ML expertise: download the release, run docker-compose up, and you have a private face recognition server whose data never leaves your machines. It is Apache 2.0 licensed and free; you pay only for your own compute. An AWS Marketplace AMI exists for one-click cloud deployment. Accuracy is whatever the bundled model you pick delivers — there are no published system-level benchmarks, which is the honest trade-off of a packaging project.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
No published benchmarks for CompreFace as a system. Accuracy depends on the bundled recognition model selected (FaceNet or InsightFace family models), whose paper benchmarks apply to the underlying models.
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: public GitHub repo by established vendor Exadel with documented REST API, SDKs (Python/JS/.NET), docker-compose one-command deploy, and AWS Marketplace listing. Strengths: Apache 2.0 license, genuinely self-hosted, clear docs. Docked for: (1) no published benchmarks of the system as a whole; (2) accuracy inherited from wrapped models rather than owned. License: Apache 2.0 (verify bundled model weights separately).
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 7.0/10
- Maintenance · recency, releases, issue load
- 7.0/10
- Openness · license + docs
- 10.0/10
- Manual review · sanity read + red-flag check
- 7.0/10
- · exadel-inc/CompreFace; docker-compose one-command self-hosted REST server (detection, recognition, verification, landmarks, age, gender, mask, head pose).
- · SDKs for Python/JS/.NET; AWS Marketplace AMI exists.
- · No published system-level benchmarks; accuracy inherited from wrapped models.