github skye-harris/hass_local_openai_llm 1.2.0

8 hours ago

Added experimental Retrieval Augmented Generation (RAG) through a local Weaviate server

  • Ability to query user inputs through a Weaviate vector database using hybrid search
  • Search results are provided to the LLM as contextual information along with the users input
  • See the README for more information

Removed Full manual prompting option: the original reason for this is no longer a thing, its not been documented, and looks to have broken in a prior update anyway

Full Changelog: 1.1.3...1.2.0

Don't miss a new hass_local_openai_llm release

NewReleases is sending notifications on new releases.