Yup. My wife has a family history of lupus, has kidney issues, had a serious b12 deficiency, and pretty much every other symptom of lupus, but a negative ANA panel, so it can’t be lupus (a negative ANA doesn’t rule it out completely).

When she went in because she was having neuropathic pain, which is very common in lupus and b12 deficiencies, she was told it was probably from her covid vaccine.

What sucks the most is I, a 6’3 male, actually gets taken seriously by the same doctors. It’s bad enough that I have to go with her to appointments so there’s a chance of her being taken seriously.

Or a doctor that dismisses all of your symptoms because they can’t be fucked to put effort into a diagnosis.

It’s a foundation model. Basically it’s the base algorithm that you train with data. LLMs are FMs that have been trained with an enormous amount of data, but they aren’t necessary for every application, especially if you only need the AI/ML to perform a specific task.

Fine tuning an FM is just feeding it your own data.

Since you clearly can’t read, I’m done discussing this with you. Maybe pick up a book and improve that reading comprehension a bit.

Oh, then that’s super simple with the media player entity.

One issue you’ll run into is that HA can’t stream music over Bluetooth (or couldn’t a few months ago), so you will have to go with a wifi speaker or preamp, or a separate server to stream the audio to a Bluetooth speaker. Or find a python package that you can incorporate into HA through the app daemon add on, or a docker image to run as an add on.

You can probably use HA to do the voice processing and send the command to the streaming server, but I haven’t played with the voice feature enough to tell you how much of a pita it would be to set it up to reliably be able to decipher a band, album, or song from a voice command. I do know that it’s a hassle to do it in rhasspy (you have to specify each option, iirc).

Once you do get it set up, you can probably set up a custom media player entity to control playback.

I’ve been checking on native HA Bluetooth streaming on a regular basis, because I want to design a whole home audio system that will follow me from room to room.

You don’t need an LLM for this. You just need a FM that you fine tune, and you’d be surprised at how little computing power is actually required.

For our uses (which are similar to what OP wants), it takes longer for us to do an OCR scan on the documents our AI works with than for Sagemaker to do it’s thing on a rather small instance.

And, devs would just be implementing API calls, so it wouldn’t be a big deal to make the switch.

Buddy, I have actual training in AI/ML from some of the leading engineers in the field, and my job leverages AI/ML very successfully to do a task really similar to what OP is looking for.

Maybe the versions available to the public to play with aren’t up to the task, but using AWS Bedrock you can absolutely get results like OP wants.

They already exist.

They’re meaning something more along the lines of an ASIC. A board specifically engineered for AI/ML.

Yup. Nvidia can’t make em fast enough to keep up with demand.

Maybe open source LLMs aren’t up to the task, but proprietary ones certainly are.

Also, you wouldn’t really need a LLM, just a FM that you fine tune for your specific purpose.

Programming AI is actually super easy, unless you decided to create your own foundation model. Even then, you would have data scientists building it, not devs.

Plenty of FMs and LLMs already exist that would be up to the task.

AI dedicated boards already exist, and Nvidia can’t produce them fast enough to keep up with demand.

Source: A senior AI engineer at AWS told me.

Yes? At the time of my comment all of the other comments were people hoping it would be a good thing.

I’m not sure why people are expecting Microsoft to act altruistically in this merger. They’re a publicly traded company that exists to create profit for their shareholders, and they’re not going to do a single thing that won’t increase their profits.

Mergers like this are always bad for the consumer, and the FTC is betraying the citizens by letting it happen.

Ah, well the qbittorrent config file should live in the directory you mapped to /config. I think you can pull the default one and edit it for your container.

I need more coffee. I read the sensor as “tuber futures” and wondered why you had a bar graph of high risk investments in HA.

Fair enough, but OPs problem can be solved by changing a 0 to a 1 in a single file.

Just use a different port number. I’m not sure why it’s necessary to use the same one since you can change qbittorrent’s port in the config files.