Update LLM API part

This commit is contained in:
Paulus Schoutsen 2024-06-04 13:02:14 +00:00
parent cabbd62ab1
commit 96a6c0e921

View File

@ -76,8 +76,8 @@ Our voice assistant brain is called a **conversation agent**.
It is responsible for understanding the intention behind the spoken command,
performing an action, and generating a response.
Since the beginning of our voice journey, we proposed you to switch your
conversation agent for a LLM-based one. This led to some very funny experiments,
Since the beginning of our voice journey, we allowed you to switch your
conversation agent for an LLM-based one. This led to some very funny experiments,
such as our [infamous interview with Mario](https://www.youtube.com/watch?v=eLx8_NAqptk).
Up until now, these two worlds (our own Home Assistant conversation agent
@ -88,7 +88,7 @@ Today we're happy to announce that we are closing this gap!
<lite-youtube videoid="3V9Pj6XCels" videotitle="Temporary video that needs to be replaced"></lite-youtube>
When setting up a LLM-Based conversation agent you can decide to let your
When setting up an LLM-Based conversation agent you can decide to let your
conversation agent control your home with a flick of a switch.
<p class='img'>
@ -96,7 +96,12 @@ conversation agent control your home with a flick of a switch.
Temporary screenshot that needs to be replaced.
</p>
This is already available on Open AI and Google AI.
This release makes this available to [OpenAI](/components/openai_conversation/) and [Google AI](/components/google_generative_ai_conversation/). To make it easier to get started with LLMs, we have updated these integrations with recommended model settings that strike the right balance between performance, speed and cost. If you had them set up previously, we recommend enabling the recommended settings and emptying the instructions.
Our recommended model settings perform equally well for voice assistant tasks. Google is 14x cheaper than OpenAI, but OpenAI is better at answering non-smart home questions. We are preparing a blog post with a deep dive into the research that went into this feature, coming soon.
_Thanks to our AI team [@shulyaka](https://github.com/shulyaka), [@tronikos](https://github.com/tronikos), [@allenporter](https://github.com/allenporter), [@synesthesiam](https://github.com/synesthesiam), [@jlpuffier](https://github.com/jlpuffier) and [@balloob](https://github.com/balloob) for this feature!_
## Assist: Media player intents