01
Xiuyu Vision-Language Model
Vision-language intelligence for video analysis, search, and proactive safety alerts.
PRODUCTS
Products & Solutions
02
Smart Campus Visual Security System
Intelligent monitoring of people, behavior, and safety events across business campuses.
03
FS Face Analytics Appliance
Upgrade existing cameras with rapid identity recognition and dynamic personnel management.
04
Industrial Visual Safety Monitoring System
Production safety monitoring for people, vehicles, equipment, and operational risks.
05
Mobile Sentinel
A portable, battery-powered intelligent monitoring solution ready to deploy anywhere.
AI Technology Shapes a Smarter Future
We specialise in Vision AI and provide consulting and development for industry-specific products and LLMs, serving clients in China and worldwide.
Clients & Partners
WHAT WE DO
Solutions & Selected Projects
VISION AI Security Solutions
We provide visual AI security solutions for factories and industrial parks in China, Western Europe, North America, Africa, and Southeast Asia, including Indonesia and the Philippines.
GLOBAL SERVICES
Consulting and Development Services for AIoT, Vision AI & LLM Industry Solutions (Overseas)
Asiantech / DentaSync
Full-stack product solutions for Asiantech Pte. Ltd.’s DentaSync platform (dentasync.xin).
European FinOps & Microsoft 365 Optimisation
Product development services for European FinOps and Microsoft 365 optimisation platforms.
Private LLM Products
Consulting and custom development for privately deployed LLM products.
Business Management & Training Systems
Custom development and consulting for internal business management and education and training systems.
Xiuyu Vision-Language Model
The Xiuyu Vision-Language Model is an intelligent analysis system built on vision-language reasoning, face and body recognition, and spatiotemporal trajectory fusion. It combines natural-language search, case management, dynamic identity clustering, body-feature capture, and watchlist alerts to reduce the cost and latency of traditional investigations across public-safety scenarios.
Smart Campus Visual Security System
Powered by AI visual analytics, this smart campus supervision system automatically captures movement trajectories, accelerates post-event searches, and detects safety risks. It replaces labor-intensive, inefficient processes with comprehensive AI-assisted security management for offices, campuses, and managed facilities.
FS Face Analytics Appliance
The FS Face Analytics Appliance connects to more than 99% of standard surveillance cameras. It automatically distinguishes employees, visitors, and unknown individuals, supports allowlist and blocklist alerts, and works across offices, campuses, schools, hospitals, factories, and other indoor or outdoor environments.
Key Features
- Works with mainstream cameras, NVRs, and video platforms without requiring new cameras.
- High capture and recognition accuracy for large angles, side profiles, low light, and other challenging conditions.
- Contactless attendance automatically records and compares attendance data without active user participation.
- Multiple face-library management distinguishes visitors, blocklists, and allowlists, with video archiving and playback.
- Dynamic face clustering consolidates capture records across multiple days to reveal movement patterns.
- Accurate indoor and outdoor recognition with support for third-party platform integration.
Industrial Visual Safety Monitoring System
This AI visual analysis system monitors production areas and lines around the clock. It detects unsafe behavior, abnormal equipment conditions, and operational risks in real time to protect people, vehicles, and assets. AI detection combined with human review enables end-to-end control of industrial safety operations.
Mobile Sentinel
Mobile Sentinel is a portable hardware-and-software monitoring solution that can run on battery power and be deployed wherever protection is needed. It captures unknown individuals, allowlisted or blocklisted faces, and human activity for continuous mobile monitoring and alerts. It supports discreet capture, online and offline operation, live mobile viewing, and important event notifications—significantly reducing the burden of on-site staffing.
Core Capabilities
Visual Recognition Algorithms
Structured detection of pedestrians, non-motorized vehicles, motor vehicles, and other target types
Re-identification (ReID), face and body clustering, and blurry-image search
Integrated training-and-inference vision-language models
Natural-language search powered by large models
Human and vehicle behavior recognition, including smoke and fire, phone use, specified apparel, and smoking
Re-identification (ReID), face and body clustering, and blurry-image search
Integrated training-and-inference vision-language models
Natural-language search powered by large models
Human and vehicle behavior recognition, including smoke and fire, phone use, specified apparel, and smoking
Custom Voice Conversation Model
Build a large model tailored to your business
Private training for large models such as Llama
Data ingestion through client applications, streaming WSS APIs, and file uploads
Support for multiple audio formats and text conversations
Switchable ASR providers, including Tencent Cloud and Alibaba Cloud, with end-to-end latency of ≤80 ms
Tenant-level configuration of algorithm capabilities
Data ingestion through client applications, streaming WSS APIs, and file uploads
Support for multiple audio formats and text conversations
Switchable ASR providers, including Tencent Cloud and Alibaba Cloud, with end-to-end latency of ≤80 ms
Tenant-level configuration of algorithm capabilities
Private LLM Solutions
Private training and deployment for large models such as Llama
Integration with healthcare and other domain-specific models
Integration of GPT, Llama, and other leading large models
Integration with healthcare and other domain-specific models
Integration of GPT, Llama, and other leading large models
Open Algorithm Platform
Our Open Algorithm Platform brings multiple AI algorithms together to deliver tailored visual intelligence for industrial operations, enterprises, and urban management. It helps organizations improve efficiency, strengthen quality control, and manage safety risks.
Platform Highlights
- Tailored Services: Algorithms can be adapted to the requirements of each industry, workflow, and operating environment.
- Fast and Accurate: Modern machine-learning techniques improve analytical accuracy while reducing false positives and missed events.
- Easy Integration: Modular APIs and compatibility with mainstream video platforms help integrators deploy visual AI quickly.
- Real-Time Analysis: Live video processing detects abnormal events quickly and supports faster incident response.
About HOUQI
Founded in 2021, Hangzhou Houqi Intelligent Technology Co., Ltd. specialises in Vision AI and consulting and development for industry-specific products and LLMs, serving clients in China and worldwide. We deliver visual security products and solutions for factories and industrial parks, alongside AIoT, Vision AI and LLM industry solution consulting and development for overseas clients.
HOUQI has focused on independent research, algorithm development, and dependable product experiences from the beginning. Our products serve customers across security, industry, and campus operations, helping them improve safety-management efficiency and unlock the value of operational data.
We continue to advance visual AI and private LLM solutions with customers and partners, turning real-world AI applications into measurable product and business value.
Business Inquiries
Company
Hangzhou Houqi Intelligent Technology Co., Ltd.
Email
cding@ihouqi.cn
Telegram
Services
Vision AI security for factories and industrial parks; industry product and LLM consulting and development; overseas AIoT, Vision AI and LLM industry solutions.
Expertise
Computer vision, speech recognition, and semantic understanding
Copyright © 2026 Hangzhou Houqi Intelligent Technology Co., Ltd.