Monitor your security cameras with locally processed AI.
- MIT
- Python
- Docker
- Actively maintained
Video Surveillance comparison
Pick Frigate for AI object detection and tight Home Assistant integration; pick ZoneMinder for a classic, mature NVR that records many cameras without needing an AI accelerator.
Both are open source and run on your own server. Below: a side-by-side of license, stack and features, live GitHub activity, and our verdict on who should pick which.
Monitor your security cameras with locally processed AI.
Closed-circuit television (CCTV) software application which supports IP, USB and Analog cameras.
Head-to-head
Hand-checked differences first, then live numbers from GitHub, refreshed with every catalog update.
| Aspect | ||
|---|---|---|
| License | MIT | GPL-2.0 |
| Language/stack | Python backend with a React web UI, shipped as a Docker image | C++, Perl and PHP, installed from distro packages or a container |
| Detection | Real-time object detection (person, car, animal) on top of motion detection | Pixel-based motion detection with zones; object detection needs add-ons such as the event notification server |
| Hardware needs | The docs recommend an AI accelerator (Coral, OpenVINO, Hailo or a GPU) for efficient detection | Runs on plain CPUs; heavy motion analysis on many streams is CPU intensive |
| Integrations | MQTT events and an official Home Assistant integration with camera entities and notifications | REST API and the third-party zmNinja app for mobile viewing and alerts |
| Learning curve | Configured mainly through a YAML file plus a config editor in the UI | Configured through a dense web console with many per-monitor options |
| GitHub stars | 36,390 | 5,941 |
| Commits, last 12 months | 1,445 | 3,316 |
| Last commit | Oct 5, 2026 | Oct 5, 2026 |
| Latest release | v0.18.0Sep 12, 2026 | 1.38.6Oct 5, 2026 |
| Main language | Python | PHP |
| Official Docker image | Yes | No |
| Repository | blakeblackshear/frigate | ZoneMinder/zoneminder |
Swipe the table sideways to see both apps. GitHub figures come from the public API. Highlighted values are the higher of the two.
Frigate is the better choice for most new home setups: it detects people, cars and animals instead of raw pixel motion, so you get far fewer false alerts, and it plugs straight into Home Assistant over MQTT. ZoneMinder suits people who want a long-established, general-purpose recorder with fine-grained zone and motion controls, or who run older hardware without a detector.
Best for
Home Assistant users who want smart person and vehicle alerts.
Best for
Traditional CCTV recording with granular motion zones on modest hardware.
Not sure yet? Spin up a small VPS or a managed instance, try both for a week, and keep the one that fits.
Sponsored links: we may earn a commission if you sign up, at no extra cost to you. It never affects rankings.
No, but the docs recommend a detector such as a Google Coral, an Intel iGPU via OpenVINO, a Hailo module or an Nvidia GPU. CPU detection works for testing but is slow and uses a lot of power.
Not out of the box. Object detection is added through community tools like the ZoneMinder event notification server with machine learning hooks, which takes extra setup.
Frigate. It has an official Home Assistant integration and publishes events over MQTT, so detections can trigger automations directly.
Frigate has 36,390 GitHub stars against 5,941 for ZoneMinder. Over the last twelve months Frigate received 1,445 commits and ZoneMinder received 3,316, so ZoneMinder is the more actively developed of the two right now. Stars measure interest, not fit, so weigh them against the differences above.