Fact · Voice and AI
Home Assistant says that before it added the prefer-local setting, many easy-to-run commands were being sent to a language model, which slowed responses and wasted tokens.
Checked against its source on October 17, 2026
In the source's own words
“we noticed many easy-to-run commands were being sent to an LLM, which can slow down things and waste tokens.”
Speech-to-Phrase brings voice home (Voice chapter 9), Home Assistant project, Home Assistant blog, 2025, Speech-to-Phrase introduction.
Open the original- What it measures
- The setting always tries the built-in agent first and passes the command to the LLM only if the built-in agent does not understand it.
- What it does not show
- A design rationale from the project. It does not give a measured time saving.