The AI Registry.
Don't be locked into proprietary, black-box vendor analytics. With OpenNVR's zero-friction plugin architecture, you can deploy your own Python models or route to community-built weights instantly.
Official Core Models
9 ActiveYOLOv8
Low-latency person and object detection on ONNX runtime — CPU and GPU. The default Tier 0 adapter.
ByteTrack
Persistent multi-object track IDs across frames. Runs as a post-processor over an upstream detector.
InsightFace
Face detection and recognition with 5-point landmarks and a REST-managed face database.
fast-plate-ocr
License-plate text recognition on a cropped plate. Designed to chain downstream of a detector.
BLIP
Vision-language scene captioning that runs locally — natural-language descriptions of a camera scene.
Open-Vocabulary Detection (OWL-ViT)
Detect whatever you can describe — "red truck", "person wearing a backpack". Free-text queries return boxes, for attributes a fixed-class detector can't express.
Whisper
Speech-to-text transcription via faster-whisper — CPU and GPU.
Piper
Fast, local neural text-to-speech for spoken alerts and the camera-agent voice loop.
Ollama LLM
Local large language models (Llama, Mistral) for video reasoning, via the bundled reference server.
Community Contributions
2 ModelsBuild & List Your Adapter
Built a custom model for a specialized industry? Don't let it sit on your drive. Plug it into the OpenNVR ecosystem and share it with the community in under 30 minutes.