# valentinfrlch/ha-llmvision

Visual intelligence for your home.

Repository: https://github.com/valentinfrlch/ha-llmvision
Canonical: https://ross.abutalabs.com/products/ha-llmvision
Homepage: https://llmvision.org
Language: Python
License: Apache-2.0
License Family: permissive
Topics: llm, vision, hacs-integration, home-assistant, ai, multimodal, cctv-detection, notifications, smart-home
Last push: 2026-08-23T06:57:15+00:00

## Health v2 (maintenance only)
Score: 90/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 96, longevity 60
- inputs: {"age_days": 841, "days_push": 10, "days_rel": 29, "gap_med": 19.0, "n_releases_24m": 21}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1440, forks 139 (observed 2026-08-28T04:04:44.246041+00:00)

## What it is
LLM Vision is a Home Assistant integration (installed via HACS) that uses multimodal large language models to analyze images, videos, live camera feeds, and Frigate events. It builds a timeline of analyzed events, updates sensors from extracted data, and supports many providers including OpenAI, Anthropic, Google Gemini, Ollama, and any OpenAI-compatible endpoint.

## Use cases
- analyze who is at my front door camera
- get notified when a package is delivered
- describe what happened in a cctv clip
- track people and pets seen by home cameras
- build a timeline of camera events in home assistant
- run local vision models on camera feeds for privacy
- create automations triggered by what a camera sees

## When to choose
- you use Home Assistant and want AI-powered analysis of cameras, images, or videos
- you want provider flexibility including fully local models via Ollama or LocalAI
- you want event timelines and sensor updates driven by visual data

## When to avoid
- you don't use Home Assistant
- you need real-time object detection without LLM inference costs or latency
- you want a standalone computer-vision pipeline outside a smart-home context

## Facets
- artifact type: plugin
- maturity: active
- function: machine-learning, computer-vision, llm-inference, image-processing, video-processing, alerting, workflow-automation
- domain: iot, artificial-intelligence, computer-vision
- platform: self-hosted, python
- tags: home-assistant, hacs, multimodal-llm, cctv, frigate, camera-analytics, timeline, home-automation

## Member repositories
- valentinfrlch/ha-llmvision (main) score 90

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:44.246041+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:36:33.501360+00:00, confidence not recorded.
  - readme: https://github.com/valentinfrlch/ha-llmvision (fetched 2026-08-28T04:04:44.246041+00:00, sha ebbf34761a90)
  - homepage: https://llmvision.org (fetched 2026-08-29T11:46:52.372232+00:00, sha dd52e19d9332)
- Data as of 2026-08-30T08:39:29.467469+00:00.