facex-engine

facex-engine / facex

Public

Face verification in the browser. 74 KB WebAssembly. No server, no cloud, no dependencies. Also runs native at 3ms on CPU.

avx-512 avx2 biometrics c99 computer-vision
47
7
100% credibility
Found Apr 26, 2026 at 47 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
C
AI Summary

FaceX is a compact library that computes face similarity embeddings locally in browsers or on servers for privacy-focused identity verification.

How It Works

1
🔍 Discover FaceX

You hear about FaceX, a super-fast way to add face matching to apps that works right on your device without sending photos anywhere.

2
📥 Get the files

Download the small package and model files to your computer with a simple script.

3
Pick your setup
🌐
Browser

Drop it into your webpage for instant face checks.

💻
Computer

Link it to your program for lightning speed.

4
🧩 Connect it

Add a few lines to recognize faces from photos and compare them easily.

5
✅ Test matches

Upload two photos and see if they match with a score, feeling the speed firsthand.

🎉 Secure logins ready

Your app now verifies faces privately and quickly, unlocking doors or apps with a smile.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 47 to 47 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is facex?

Facex delivers face verification straight in the browser via a 74 KB WebAssembly module or natively in C at 3ms per face on CPU—no servers, clouds, or dependencies needed. It takes a 112x112 RGB photo, computes a 512-dim embedding, and compares pairs via cosine similarity (above 0.3 signals a match), hitting 99.73% accuracy on LFW benchmarks. Developers drop it into apps for quick identity checks without shipping photos anywhere.

Why is it gaining traction?

Its zero-footprint design crushes alternatives like ONNX Runtime (1.3x slower, 200x larger) or InsightFace, running offline at browser speeds of 17ms. No Python runtime or GPU required—just link the static lib or load the WASM SDK for instant face verification in web apps. The CLI and Go bindings make prototyping painless, ideal for dodging "face verification failed" errors on platforms like Binance or GCash.

Who should use this?

Frontend devs building face login for Instagram, Roblox, or Facebook-style apps where users hate server roundtrips. Edge hardware folks for access control gates or video door stations like Facexess. Proctoring tools verifying exam takers via webcam without cloud dependency.

Verdict

Grab it for client-side face verification prototypes—solid API and benchmarks make it a no-brainer for low-latency needs. With just 47 stars and 1.0% credibility, it's early-stage (basic docs, no tests shown), so pair with a detector like SCRFD for production.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.