2024.6: Dipping our toes in the world of AI using LLMs 🤖

Again, another amazing release from the HA team.

I have to say A BIG THANK YOU!! to @karwosts for developing collapsible sections in blueprints. I get asked about this all the time, and seeing it go from a feature request to being developed and now implemented… nice. As always, I have tested it, and it works flawlessly. I will be rolling this out in many of my blueprints. It makes them so clean and very easy to use now. On behalf of everyone using blueprints and all the blueprint developers, THANK YOU!!

Watching the YouTube release… JLo, you are a weapon.

Blacky :smiley:


Wow, what an amazing release! Thanks so much to everyone who’s been working on this - you are all epic. :smiley:

How complicated is it to add the autocorrect to home assistant

Could something like a Coral TPU enable a RPi to run a local model?

Since upgrading to 2024.6 I am experiencing many issues/outages.
I am new to Home Assistant but not networking or home automation.

So far, Govee, Shelly, Sonos, Nanoleaf are all failing, plus not sure if related but the Symfonisk Remote Blueprint is configuring but unable to see any of the buttons presses visible in Home Assistant Device page and the Automation is not showing either.

Not sure if these are related yet.

Need to figure out how to roll back to 2024.5.

Looks like I just run ha core update --version 2024.5.6 ?

the ha core update --version 2024.5.6 command will revert you but not solve the issues…

have a look here How to help us help you - or How to ask a good question, post your logs, share what config and setup you have and who knows maybe the issues(s) you have can be solved…you’ll never know if you don’t ask :grin:

1 Like

Did you miss this bit?

Local LLMs have been supported via the Ollama integration since Home Assistant 2024.4. Ollama and the major open source LLM models are not tuned for tool calling, so this has to be built from scratch and was not done in time for this release. We’re collaborating with NVIDIA to get this working – [they showed a prototype last week.] (https://youtu.be/aq7QS9AtwE8?si=yZilHo4uDUCAQiqN)


That appears to have been added after publication and after my comment, as this snapshot from the Wayback Machine shows. You can compare the timestamp of my comment with that of the Wayback snapshot: the passage wasn’t there when I commented.


Well, I guess they sort-of responded to your comment, then.


If you want to use Gemini in Europe, there is no free plan available. But, if you activate first Gemini free credit in a project in Google Cloud console, you will get three months of free Gemini in the Gemini studio. Then activate Gemini API through the studio console, and here you are, 3 months of Gemini credit.

First try are quite awesome, despite some issues with special characters in room names, like Cheminée in French.

But so cool to ask Assist : What is the warmest room ? Open covers in the kitchen half way …

So so cool. Sure I would prefer to have it running a local LLM but it a first approach to replace Siri in day to day use.

1 Like

Good morning all. It’s a Saturday here and my weekday routines have all fired! Seems something is up with workday_sensor

1 Like

Two days after 2024.6 and there’s already an update?

This pace of software releases is too fast. I’d prefer stability to new features.

How about even months there’s UI updates and odd months there’s integrations updates?


Yep, I’ve just discovered the same thing. My workday sensor is ‘on’ despite being Saturday.

1 Like

This workaround automation seems to handle it for me.

alias: Workday sensor fix
  - platform: time
    at: "00:00:00"
  - service: homeassistant.reload_config_entry
        - binary_sensor.workday_sensor
        - binary_sensor.workday_tomorrow
        - binary_sensor.workday_yesterday

1 Like

I think we’ll have to agree to disagree. I think that modern generative LLMs do not understand the words they output. They work based on word distributions following the trained statistical path through their network. But I will acknowledge that for simple queries they seem able to predict the intent of the query and produce valuable results. It seems to me that Home Assistant’s voice is a good candidate for simple AI use.
In fact the company Hailo recently partnered with Raspberry Pi to produce a 70$ AI kit with a neural network M.2 card for computer vision acceleration. And Hailo has announced a forthcoming processor able to process LLMs. This might be another local solution to language processing in Home Assistant.
If you haven’t read it yet, this blog post about HA and the furture of AI interfaces is very interesting (and reassuring to me).

1 Like

Having a non-YAML background control is great! Now, I wish there was a card transparency control. And if there was a “frosted glass” effect — that would be even better.


No one is forcing you to update. There are plenty of people who only update annually. You apparently don’t know the difference from open source and production software. If you want stable production-quality IOT that rarely updates, then buy a Bosch or Seimens system. But don’t ask them for DIY help or adding hardware from other manufacturers. But, you will never be bothered with monthly updates.

In my home I have over 110 WiFi clients. Many are things I built using ESP modules. I like the flexibility that Home Assistant allows. If all you want to do is turn some lights on and off and never be bothered with an update, then buy one of those very expensive closed systems.


FYI, understand that ”Y” releases in ”YEAR.X.Y” only add cherry-picked bug-fixes, not new features, as it is only the ”X” releases that adds new features.

Example: 2024.6.1 and so on only adds cheery-picked bug-fixes to the 2024.6.0 release.

Regardless, stating that there are too many releases is a stupid thing to complain about to a free and open-source project.

Be glad that there are releases at all and simply choose not to update each release, (I know many that only update once a month, usually at the end of the month).

Yes, that was the problem. Bus error. I did not want to put dev versions on my main controller so I was stuck to 2024.5.5.

i noticed it being detected and discussed before my post.
Now it is available in 2024.6.1 and my system works again.

thanks for the great work!

You can got to Settings → Voice Assistants → Exposed Entities → Expose Entities …



Then you have a list of checkboxes