valentinfrlch/ha-llmvision
A Home Assistant integration leveraging multimodal LLMs to analyze surveillance camera feeds and events with AI-powered visual understanding.
Not currently ranked — collecting fresh signals.
star history
This project integrates multimodal large language models into Home Assistant for analyzing images, videos, live camera feeds, and Frigate events. It supports numerous LLM providers including OpenAI, Anthropic, Gemini, Ollama, and any OpenAI-compatible endpoint. The system can answer questions about visual content, remember people and objects across sessions, and maintain a searchable timeline of analyzed camera events with customizable prompts and notifications.
Frequently asked
- What is valentinfrlch/ha-llmvision?
- A Home Assistant integration leveraging multimodal LLMs to analyze surveillance camera feeds and events with AI-powered visual understanding.
- Is ha-llmvision open source?
- Yes — valentinfrlch/ha-llmvision is open source, released under the Apache-2.0 license.
- What language is ha-llmvision written in?
- valentinfrlch/ha-llmvision is primarily written in Python.
- How popular is ha-llmvision?
- valentinfrlch/ha-llmvision has 1.4k stars on GitHub.
- Where can I find ha-llmvision?
- valentinfrlch/ha-llmvision is on GitHub at https://github.com/valentinfrlch/ha-llmvision.