frigate
Open‑source NVR with real‑time AI object detection for IP cameras
Best for
DIY hobbyists who want local AI video monitoring on a Raspberry Pi
Skip if
Users needing plug‑and‑play mobile UI or cloud backup
What is frigate?
Frigate is a self‑hosted network video recorder that leverages TensorFlow Lite to provide on‑premise, real‑time object detection for any IP camera. It integrates tightly with Home Assistant and offers a lightweight, Docker‑based deployment that runs on modest hardware like a Raspberry Pi. The system stores video clips locally, generates searchable events, and provides a responsive web UI for live view, playback, and configuration. Because all processing happens locally, privacy is preserved and no cloud subscription is required.
SpecificationsAI-estimated
Key Features of frigate
Use Cases for frigate
Home Security
Monitor entrances and receive instant alerts when people or vehicles are detected.
Pet Monitoring
Track pet movement and receive notifications when they enter restricted zones.
Smart Home Automation
Trigger Home Assistant automations based on detected objects, such as turning on lights when a person arrives.
DIY Surveillance for Small Businesses
Record and review events without a costly cloud subscription, keeping footage on‑premise.
Pros & Cons of frigate
Pros
- Fully free and open source.
- Runs locally, preserving privacy.
- Low hardware requirements.
- Deep integration with Home Assistant.
Cons
- Requires technical knowledge to set up Docker.
- Limited official UI for mobile devices.
- No built‑in cloud backup; must be added manually.
Frequently Asked Questions
Does Frigate require a cloud service?
No, all video processing and storage are performed locally on your own hardware.
Can I use Frigate with my existing Home Assistant setup?
Yes, Frigate integrates via MQTT and provides entities that can be added to Home Assistant dashboards.
What hardware is recommended?
A Raspberry Pi 4 with 4 GB RAM or any x86_64 Linux machine with Docker; GPU acceleration improves performance.
Is it possible to add custom AI models?
Yes, you can upload your own TensorFlow Lite models to detect specific objects.
Pricing Overview
View full pricing →Detailed plans are not listed. Visit the official website for pricing information.
Reviews & Ratings0.0
No reviews yet. Be the first to write one!
Top Alternatives & Similar Tools
View all alternatives & similar tools →No alternatives available yet.
People also viewed
Related searches
About the Tool
Is this your tool?
Claim this page to update details, reply to user reviews, and drive more traffic to your product.
Claim this Product →