Home-LLM

Home-LLM connects Home Assistant to language models for voice and chat control of smart devices. It's for Home Assistant users who want those requests handled on their own hardware.

• Smart home control: Give natural-language commands for lights, locks, climate controls, media players, vacuums and other supported devices.
• Voice and chat: Use the Home Assistant chat interface or its speech-to-text and text-to-speech pipelines, including streaming voice responses.
• Model choice: Use the home-focused Home-Llama-3.2-3B and Home-FunctionGemma-270m models, or instruction-tuned models such as Qwen3, Llama 3 and Mistral.
• AI automations: Extract data and generate structured responses or dynamic content through Home Assistant AI tasks.
• Language support: Built-in support covers English, German, French, Spanish and Polish.

Models can run inside Home Assistant through llama.cpp or on another machine through Ollama, llama.cpp server, LM Studio, LocalAI, vLLM or text-generation-webui. The small Home models can run on a Raspberry Pi without a GPU. Local use doesn't require a cloud service. Optional OpenAI Responses and Anthropic Messages connections send data to the provider, including the state of exposed devices and potentially your location.