Does mtp really improve t/s?

Try subwave. You can have AI djs choose music from your navidrone library based on shows you create by defining prompts, moods and styles.

Pretty neat…

Fast developing and pretty solid already.

Given how lineage os works (was a maintainer in the past), no.

A new device even if 95% similar will not work with older lineage builds for other devices.

You need the other device build.to explicitly add support for the newer model or wait for a maintainer to create a specific build for the new phone.

Of course, non official builds will exist long time before they become official (if ever) so I would still buy the new one and wait.

No. To me, comments and engagement is what keeps me there.

Just a stream of empty posts meh.

I think your personal bias is obfuscating your understanding of my post.

Are you assuming cars are a positive development unlikely bombs or guns? None of them are. I was not making positive examples, exactly as I am not defending AI as it stands today.

You got it all heads and tails my friend. Go read it again.

Guys, I am talking to all the zealots people who loudly open their mouths to talk down anything referring to AI.

AI is here to stay. Like cars after horses. Like guns after swords. Like h-bombs after mortars. The sooner we regulate it, learn how to use it ethically and safely the better.

Just asking for it to be deleted or forgotten has never worked in human history and never will. As well as blindly criticizing whoever uses it for any reason has never worked out and, again, won’t work.

Let’s regulate it. Let’s fight for an ethical use instead.

Things usually break on updates unless your disk fills up or worse some hardware fails…

So it’s pretty easy to fix: never upgrade automatically.

I always upgrade manually and check immediately what I have upgraded. Works fine so far.

And I was only partially joking… Yeah if a service stays broken for a long time, just uninstall it, it was useless.

Observawhat? My services just exist, if they fail and nobody complains… Well, they get removed. If the fail and somebody complains… The get restarted…

Pretty efficient but simple approach.

20/30 Is double what my two RTX A4000 archives… So feels good :)

On which hardware do you run qwen 3.8?

I thought ai was banned on codeberg

This. That’s like vibe coding. Vibe arting get you slop as well.

Now, give an AI to a good coder, and you don’t get slop code. Same with artists

Do you still get AI stuff? Yes, but a whole better result.

Of course, Gentoo Linux! :)

In my limited experience, AI properly guided can be a great bug hunter.

As usual with AI anything must be thoroughly vetted of course, but overall it’s a huge time saver.

I believe that it’s impossible to avoid ai generated or ai committed or ai designed or ai helped code. Or will very soon be.

Better to learn to cope with it.

It is possible with nvidia-smi, I recently found out about. I need to esperiment …

Mmm… Nah, i fully disagree.

Hobby doesnt mean grindy. I very much prefer my power tools when woodworking, to manually saw, dust and chisel wood…

Same thing here, using AI is the power tools of woodworking.

I would never build furniture by manually sawing boards. Very much prefer to power cut and save that grind time to paint a second layer or build an extra piece.

Really, the boring grind adds nothing. Now, there is grind that adds value by taking time, and sometimes it’s what you look for into hobbies. Time consume.

Just, sometimes not.

Yeah i managed to acquire two 16gb Nvidia cards and built a Ryzen box around it. With 32gb vram and llamacpp you can do wonders . Maybe slowly.

Anyway the cards alone are over 2k€ nowadays, and they are OLD, so really crazy.

And the power consumption… My rig is round 500w when operating with full GPUs… So can get quite expensive quickly. Easy to get to 5kw per day with only a few hours of llm running.

Is this just plain advertising? I think it should be deleted.

Qwen 3.8 27b is a beast. Better quality than all the free tier commercial AIs. If you can afford a big enough context window, it can very well assist you as good as paid tier commercial ai.

It’s the first model I feel confident to assign a complex coding task and have it actually deliver a final product.

On my hardware it’s pretty slow (15t/s) but also quite usable provided you are not in a hurry.

I’m about 3h ingested a complex software and added a full ffmpeg pipeline to deliver video transcoding to the browser with backend side caching and more.

It would have taken me days just to dig into how to use ffmpeg for that taks, and more just to debug it.

Even hobbies have grind parts that you might as well get some help on.