I agree generally that on Mac you can 'get by' with 8gb and for the target audience on this, and how they'll likely use it - it's totally acceptable.
But if it's for serious work, this is not the device. 'Managing' the software to 'use the machine well' to get serious work done is unacceptable in 2026. It needs to just work and disappear into the background. I have enough to think about and micro managing the software running is out of the question.
> 'Managing' the software to 'use the machine well' to get serious work done is unacceptable in 2026
I agree, I just don't think the rush to get more and more RAM and storage is the root of the problem.
Why on earth does a browser need more than 10 GB to display web pages?? Why does macOS keep piling/hiding trash that should be deleted in "System Data"?
And, if you need to keep device backups, put them on an external drive; that's what those things are for.
It depends on how you define "serious work". Is it to get the best results possible, or is it to tax a computer as much as possible? Programmers would usually answer the latter, while users would answer the former.
That's why programmers put their stuff into Kubernetes which go into virtual machines, which go into eleven layers of javascript abstraction which go into twelve thousand node packages, which go into something else to end up with something with very basic functionality, which usually doesn't work very well.
Other pro computer users are focused on the results, so they use professional office software, calendars, communications, photo and video editing and effects, photo-realistic 3D editors, studio level audio and music editing software. All which lives perfectly fine on 8GB of RAM.
An “intent” is something that a person wants to do - set a timer, get directions, etc.
A “slot” is the variable part of an intent. For instance “I want directions to 555 MockingBird Lane”. Would trigger a Directions intent that required where you are coming from and where you are going. Of course in that case it would assume your location.
Back in the pre LLM days and the way that Siri still works, someone had to manually list all of the different “utterances” that should trigger the intent - “Take me to {x}”,”I want to go to {x}” in every supported language and then had to have follow up phrases if someone just said something like “I need directions” to ask them something like “Where are you trying to go”.
Now you can do that with an LLM and some prompting and the LLM will keep going back and forth until all of the slots are filled and then tell it to create a JSON response when it has all of the information your API needs and you call your API.
This us what a prompt would look like to use a book a flight tool.
Using LLMs for voice assistants is relatively new at scale that’s the difference between Alexa and Alexa+ and Gemini powered Google Assistant and what Apple has been trying to do with Siri for two years.
It’s really just using LLMs for tool calling. It is just call centers were mostly built before the age of LLMs and companies are slow to update
Understood. This overlaps with a side project where I’m getting acceptable (but not polished) results, so trying to do some digging about optimizations. Thanks!
One of my niches is Amazon Connect - the AWS version of Amazon’s internal call center. It uses Amazon Lex for voice to text. Amazon Lex is still the same old intent based system I mentioned. If it doesn’t find an intent, it goes to the “FallbackIntent” and you can get the text transcription from there and feed it into a Lambda and from the Lambda call a Bedrock hosted LLM. I have found that Nova Lite is the fastest LLM. It’s much faster than Anthropic or any of the other hosted ones.
It's an absolutely brilliant bit of maths that breaks a complex waveform into the individual components. Kind of like taking an orchestral song and then working out each individual instrument's contribution. Learning about this left me honestly aghast and in shock that it's not only possible but that someone (Joseph Fourier) figured it out and then shared it with the world.
Some hobbies also happen to be exercise and can be done nearly daily. Just do more of them and now there's a much lower 'cost' to exercise. Whether or not you find these types of hobbies enjoyable is another story.
I'd go for the A770 over the B580. 16GB > 12GB, and that makes a difference for a lot of AI workloads.
An older 3060 12GB is also a better option than the B580. It runs around $280, and has much better compatibility (and, likely, better performance).
What I'd love to see on all of these are specs on idle power. I don't mind the 5090 approaching a gigawatt peak, but I want to know what it's doing the rest of the time sitting under my desk when I just have a few windows open and am typing a document.
All the touch screen does is make it top heavy and the hinge less effective at damping the movement.