Face Analysis
Services & Licensing
for Enterprise
InsightFace provides world-leading face analysis solutions — from face recognition and face swapping model licensing to deepfake detection, Live Identity for streaming, cross-platform SDK, and custom AI services. Deploy state-of-the-art face AI in your product with commercial confidence.
1import insightface2import cv234# Face Recognition5app = insightface.app.FaceAnalysis('buffalo_l')6app.prepare(ctx_id=0)7img = cv2.imread('photo.jpg')8faces = app.get(img)9print(faces[0].embedding.shape)1011# Face Swapping12swapper = insightface.model_zoo.get_model(13 'inswapper_128.onnx')14res = swapper.get(img, faces[0], faces[1])15cv2.imwrite('swapped.jpg', res)Choose Your Path
Five ways to work with InsightFace
Choose the right path for your team — from licensed recognition models and private evaluation to authorized live identity, face swapping, and deepfake detection.
Face Recognition Licensing
License high-performance InsightFace recognition models and the cross-platform InspireFace SDK — with built-in liveness detection — for commercial products, identity verification, access control, and on-device or private deployment.
Private Model Evaluation
Benchmark private recognition models with your own validation data through a structured enterprise evaluation process.
Live Identity
NEWRun authorized real-time AI identity transformation locally for professional live streaming and virtual production workflows.
Face Swap API & Licensing
Access reviewed commercial face swap APIs or licensing options for approved creative, entertainment, and product use cases.
Deepfake Detection
Detect AI-generated faces, face swaps, and forged media with enterprise deepfake detection — try the online demo in your browser, then deploy via SDK and API for KYC, content moderation, and trust & safety.
Trusted by teams building the next generation of face AI
Solutions
Everything you need for
production face AI
Comprehensive face analysis solutions — from model licensing and face swapping to cross-platform SDK and deepfake detection. Each backed by state-of-the-art peer-reviewed research.
Face Recognition Model Licensing
Commercial licensing for InsightFace open-source model packages (buffalo_l, antelopev2, buffalo_s, buffalo_m, etc.) and proprietary high-accuracy models. From research to production deployment with full legal compliance.
Read DetailLive Identity for Streaming
NEWRun authorized real-time AI identity transformation locally on a streamer’s NVIDIA GPU for professional live streaming and virtual production workflows.
Read DetailFace Swapping
Open-source InSwapper-128 has become the de facto standard for open-source face swapping, with commercial licensing, high-accuracy private models, mobile real-time models, and a local NVIDIA GPU real-time solution for professional live streaming.
Read DetailDeepfake Detection
Enterprise deepfake and face-forgery detection via SDK and API. Try the online demo to identify AI-generated faces, face swaps, and manipulated media with calibrated risk scoring for KYC, content moderation, and trust & safety.
Read DetailInspireFace SDK
Cross-platform C/C++ face recognition SDK with built-in anti-spoofing and liveness detection. Commercial licensing with dedicated technical support for Linux, Android, iOS, macOS, and embedded devices.
Read DetailCase Studies
Built for Enterprise Face AI Deployment
From identity verification and access control to creative AI and edge deployment, InsightFace helps teams evaluate, license, and deploy face AI models responsibly.
Global Fintech KYC
A fintech team evaluating face recognition models for identity verification, threshold tuning, and production deployment.
Edge Access Control
An access control provider exploring lightweight face analysis SDK deployment across embedded and edge devices.
Creative Face AI Application
A consumer application team reviewing face swap API access, licensing requirements, and responsible use controls.
Anonymized scenarios derived from typical enterprise engagements. Specific customer names are not disclosed for confidentiality reasons.
Licensing & Pricing
License models, ship with confidence
From open-source research to model licenses, local live identity deployment, production APIs, and SDK access — choose the licensing option that fits your business needs.
Open Source
For research and non-commercial projects
- Code released under MIT License — free for commercial use
- Pre-trained models for non-commercial academic research only
- Commercial use of models requires contacting us for a commercial license
- Python package (insightface) available on PyPI
- Community support via GitHub Issues
- Web demos access
Model Licensing
Commercial licensing for face recognition and face swapping models
- Face recognition model commercial license (buffalo_l, antelopev2, buffalo_s, etc.)
- InSwapper-128 face swapping commercial license
- Deepfake / face-forgery detection model commercial license
- Closed-source high-accuracy models available
- Production deployment rights
- Email support
- License compliance guidance
Enterprise & API
Full suite with API access, SDK, and dedicated support
- Everything in Model Licensing
- High-accuracy face swap API access
- Mobile real-time face swap model access
- InsightFace Live Identity organization licensing for local NVIDIA GPU streaming workflows
- InspireFace C/C++ SDK commercial license
- Deepfake detection SDK & API access
- Priority SLA & dedicated engineering
- Custom model training & AI consulting
Integration
Ship in minutes, not months
Whether you prefer Python, C++, or REST — InsightFace integrates into your stack with minimal friction.
Python Package
pip install insightface — start analyzing faces in 3 lines of code.
pip install insightface onnxruntime-gpu
# That's it. You're ready.
from insightface.app import FaceAnalysis
app = FaceAnalysis('buffalo_l')
app.prepare(ctx_id=0)InspireFace C/C++ SDK
Native SDK for mobile and embedded devices. One API across all platforms.
// Initialize the SDK
HResult ret = HFLaunchInspireFace(path);
// Create a session
HFSession session;
HFCreateInspireFaceSession(&session, mode);
// Detect and recognize
HFDetectFaces(session, image, &faces);
HFExtractFeature(session, image, face, &feature);REST API
Enterprise-grade REST API service supporting high-precision face swapping models and high-precision private face recognition models. Integrate face AI via simple HTTP calls.
# Make an API call
curl -X POST https://api.insightface.ai/extract \
-H "Authorization: Bearer <YOUR_TOKEN>" \
-F "image=@photo.jpg" \
-F "return_face_data=true"Research & Publications
Backed by world-class research
Published at top-tier academic venues. Every algorithm in InsightFace is peer-reviewed and reproducible.
ArcFace: Additive Angular Margin Loss
CVPR 2019
The foundational loss function for face recognition, setting new benchmarks on all major datasets.
RetinaFace: Single-Shot Multi-Level Face Localisation
CVPR 2020
Practical single-stage face detector with joint face detection and landmark localization.
SCRFD: Sample & Computation Redistribution for Efficient Face Detection
ICLR 2022
Efficient high-accuracy face detection with NAS-based architecture search.
Partial FC: Training 10 Million Identities on a Single Machine
CVPR 2022
Scalable training strategy enabling face recognition with massive identity pools.
Sub-center ArcFace: Boosting Face Recognition by Large-scale Noisy Web Faces
ECCV 2020
Handling noisy training data by learning multiple sub-centers per class.
Achievements
Competition winners
NIST FRVT
Top Performer on VISA and Wild Tracks
ECCV 2022 WCPA — Rank #1
3D Face Reconstruction Challenge
ICCV 2021 MFR Challenge
Masked Face Recognition Workshop
Open Source Codebase
InsightFace's code is released under the MIT License with no limitations for commercial usage. Pre-trained models are for non-commercial research only — contact us for commercial model licensing.
View on GitHubBlog
Latest Insights
Stay up to date with the latest developments in face AI technology, research, and applications.
June 2026 Face Swapping Papers: Conversational Talking Faces, Fast Portrait Animation, and Privacy Protection
A June 2026 arXiv briefing on face swapping and talking-face research, covering interactive avatars, diffusion caching, and proactive face-swap protection.
June 2026 Face Recognition Papers: Low-Resolution MoE, Efficient ViTs, and 1024-Byte Travel Documents
A June 2026 arXiv briefing on face recognition research covering low-resolution recognition, early-exit ViTs, and ultra-compact biometric document images.
June 2026 Face Detection Papers: Fairness Benchmarks, Neonatal Detection, and PAD Bias
A June 2026 arXiv briefing on face detection research, covering demographic annotations, neonatal clinical detection, and architectural fairness in presentation attack detection.
Ready to deploy the world's best face AI?
Whether you need face recognition model licensing, face swapping commercial rights, a live streaming identity deployment, a local real-time identity workflow, InspireFace SDK, or custom AI cooperation — contact us to start the conversation.
We review commercial, enterprise, and research cooperation requests. Please use a company email address when possible.