Shimitar

Yeah problem is when upgrades break stuff…

Unattended upgrade?

Why? Specially on critical systems?

If you do such upgrades, schedule for a time you don’t use the system at very least …

Reverse proxy my friend…

Using one static IP per service feels so much like … 1990s…

… As well as owing a /24, I admit.

Not really, they are testing your passwords against have I been pawnd. I agree that transparency should apply here, and also an upfront question to skip.

But it’s not “unsafe” or a privacy breach per se. No password has actually been “sent” anywhere only a partial hash that is useless for accessing your home assistant.

Yes i do.

I esperiment a lot, but my stable setup is Qwen 3.8 27B Q6_K on my home server. Hardware is Ryzen 9 5950x with 64gb ddr4 and two old but effective NVIDIA RTX A4000 for a total of 32gb VRAM.

Qwen 3.8 is the best model I can run in terms of output quality. It can perform all tasks pretty well and it’s reasonably fast at 500ts prefill and 30/40ts processing. There are much faster models but honestly not as good.

I can also run gemma4 all on CPU, with no GPU at all.

The key point is having both GPU on pci-e CPU, not on chip set bridge, so I am using a PCI riser 8x8x to share the 16x slot between the two GPUs. Inference speed is unaffected, but prefill benefits immensely.

Software side, use llama.cpp on Linux, you get the best performance possible at the expense of a bit more complexity compared to other engines.

From a power consumption perspective, I tined down my CPU governor and maximum frequency, and also the GPUs maximum watts. At ide the system is around 80w. At peak, it’s about 280w with the limiters, and inference speed doesn’t suffer too much. Without limiters I can easily reach 650w.

Edit: I run llama-swap in front of llama-cpp, and openwebui as agent manager/ web GUI. I also have ComfyUI installed because why not but seldomly use it.

You can read more of my setup on my wiki https://wiki.gardiol.org/8-llm-and-ai (no ads, no monetization whatsoever because fuck a hobby should not be a side gig)

I bought it new, but of course you can repurpose old hardware!

I bought a firewall appliance based on an Intel CPU, with 4x 2.5gbps Ethernet ports, 8gb ram. Installed opnSense on it and been super happy even since!

A bit pricey the hardware, but it’s been totally rock solid for almost two years now.

I have configured unbound with ad blocking on opnSense, and it has every single option I would ever want on it. Amazing stuff.

What do you mean?

Who chooses the tracks?

Anyway yes, set a muted dj who never speak.

And create playlists and only play those. This will bypass any AI usage.

Subwave. A bit on the AI side… But it’s brilliant and really works great.

Only problem, you want to keep that private ad it’s most probably illegal anywhere to broadcast publicly without a license. Even on the internet

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.

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.

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.

Working on it… Good suggestion.

I don’t selfhost secure, is this a new service? I selfhost securo instead, great tool!

Am I missing something here?

That sucks… Maybe a good update to LazyNVR would let grab the RTSP stream…

Good! I was probably doing something wrong, but in any case seems more logical to leverage the cameras internal detection.

Feels like nextcloud…

Have you read the readme? There is literally a text config file to fill with the camera info (url of image).

Feel free to open a ticket on codeberg or DM/PM me for any additional detail.

Indeed documentation could be better…

In my experience frigate is quite CPU intensive. I have 10 cameras and my old I7-8700 would die trying to have frigate process them all.

I opted for a different approach with LazyNVR (https://codeberg.org/LazyNVR/lazynvr-sources) leveraging the capability of the camera to perform motion detection and video recording directly. This offloaded my CPU and now I can record all 10 cameras on a OPi3.