nagaram
@nagaram@startrek.website
- Comment on That one Pokémon 2 days ago:
Softshell turtle
- Comment on 1U mini PC for AI? 3 days ago:
Honestly if you’re not gaming or playing with new hardware, there is absolutely no point.
I’ve considered swapping this computer over to Fedora for a hot minute, but it really is a gaming PC and I should stop trying to break it.
- Comment on 1U mini PC for AI? 3 days ago:
True, but I have an addiction and that’s buying stuff to cope with all the drawbacks of late stage capitalism.
I am but a consumer who must be given reasons to consume.
- Comment on 1U mini PC for AI? 3 days ago:
The Lenovo Thinkcentre M715q were $400 total after upgrades. I fortunately had 3 32 GB kits of ram from my work’s e-waste bin but if I had to add those it would probably be $550 ish The rack was $120 from 52pi I bought 2 extra 10in shelves for $25 each the Pi cluster rack was also $50 (shit I thought it was $20. Not worth) Patch Panel was $20 There’s a UPS that was $80 And the switch was $80
So in total I spent $800 on this set up
To fully replicate from scratch you would need to spend $160 on raspberry pis and probably $20 on cables
So $1000 theoratically
- Comment on 1U mini PC for AI? 4 days ago:
Ollama and all that runs on it its just the firewall rules and opening it up to my network that’s the issue.
I cannot get ufw, iptables, or anything like that running on it. So I usually just ssh into the PC and do a CLI only interaction. Which is mostly fine.
I want to use OpenWebUI so I can feed it notes and books as context, but I need the API which isn’t open on my network.
- Comment on 1U mini PC for AI? 5 days ago:
I was thinking about that now that I have Mac Minis on the mind. I might even just set a mac mini on top next to the modem.
- Comment on 1U mini PC for AI? 5 days ago:
Ollama + Gemma/Deepseek is a great start. I have only ran AI on my AMD 6600XT and that wasn’t great and everything that I know is that AMD is fine for gaming AI tasks these days and not really LLM or Gen AI tasks.
A RTX 3060 12gb is the easiest and best self hosted option in my opinion. New for >$300 and used even less. However, I was running with a Geforce 1660 ti for a while and thats >$100
- Comment on 1U mini PC for AI? 5 days ago:
A mac is a very funny and objectively correct option
- Comment on 1U mini PC for AI? 5 days ago:
I think I’m going to have a harder time fitting a threadripper in my 10 inch rack than I am getting any GPU in there.
- Comment on 1U mini PC for AI? 5 days ago:
I do already have a NAS. It’s in another box in my office.
I was considering replacing the PIs with a BOD and passing that through to one of my boxes via USB and virtualizing something. I compromised by putting 2tb Sata SSDs in each box to use for database stuff and then backing that up to the spinning rust in the other room.
How do I do that? Good question. I take suggestions.
- Comment on 1U mini PC for AI? 5 days ago:
With a RTX 3060 12gb, I have been perfectly happy with the quality and speed of the responses. It’s much slower than my 5060ti which I think is the sweet spot for text based LLM tasks. A larger context window provided by more vram or a web based AI is cool and useful, but I haven’t found the need to do that yet in my use case.
As you may have guessed, I can’t fit a 3060 in this rack. That’s in a different server that houses my NAS. I have done AI on my 2018 Epyc server CPU and its just not usable. Even with 109gb of ram, not usable. Even clustered, I wouldn’t try running anything on these machines. They are for docker containers and minecraft servers. Jeff Geerling probably has a video on trying to run an AI on a bunch of Raspberry Pis. I just saw his video using Ryzen AI Strix boards and that was ass compared to my 3060.
But to my use case, I am just asking AI to generate simple scripts based on manuals I feed it or some sort of writing task. I either get it to take my notes on a topic and make an outline that makes sense and I fill it in or I feed it finished writings and ask for grammatical or tone fixes. Thats fucking it and it boggles my mind that anyone is doing anything more intensive then that. I am not training anything and 12gb VRAM is plenty if I wanna feed like 10-100 pages of context. Would it be better with a 4090? Probably, but for my uses I haven’t noticed a difference in quality between my local LLM and the web based stuff.
- Comment on 1U mini PC for AI? 5 days ago:
That’s fair and justified. I have the label maker right now in my hands. I can fix this at any moment and yet I choose not to.
I’m man feeding orphans to the orphan crushing machine. I can stop this at any moment.
- Comment on 1U mini PC for AI? 5 days ago:
Oh and my home office set up uses Tiny in One monitors so I configured these by plugging them into my monitor which was sick.
I’m a huge fan of this all in one idea that is upgradable.
- Comment on 1U mini PC for AI? 5 days ago:
These are M715q Thinkcentres with a Ryzen Pro 5 2400GE
- Comment on 1U mini PC for AI? 5 days ago:
Not really a lot of thought went into rack choice. I wanted something smaller and more powerful than my several optiplexs I had.
I also decided I didn’t want storage to happen here anymore because I am stupid and only knew how to pass through disks for Truenas. So I had 4 truenas servers on my network and I hated it.
This was just what I wanted at a price I was good with at Like $120. There’s a 3D printable version but I wasn’t interested in that. I do want to 3D print racks and I want to make my own custom ones for the Pis to save space.
But this set up is way cheaper if you have a printer and some patience.
- Comment on 1U mini PC for AI? 5 days ago:
Not much. As much as I like LLMs, I don’t trust them for more than rubber duck duty.
Eventually I want to have a Copilot at Home set up where I can feed a notes database and whatever manuals and books I’ve read so it can draw from that when I ask it questions.
The problem is my best GPU is my gaming GPU a 5060ti and its in a Bazzite gaming PC so its hard to get the AI out of it because of Bazzite’s “No I won’t let you break your computer” philosophy, which is why I did it. And my second best GPU is a 3060 12GB which is really good, but if I made a dedicated AI server, I’d want it to be better than my current server.
- Submitted 5 days ago to selfhosted@lemmy.world | 62 comments
- Comment on Single board computer for selfhosting 6 days ago:
I just built a mini rack with 3 Thinkcentre tiny PCs I bought for $175 (USD) on eBay. All work great.
- Comment on how good are you at lying during job interviews? 1 week ago:
I always tell the interviewer what they want to hear. Its very obviously a game of correct answers.
I lie on my resume too, but not in ways I can’t back up.
For instance, I imply I have a degree because I did go to college for 4 consecutive years for a multitude of degrees. So I have different resumes with different majors depending on what job I’m applying to. I mostly use my CS/Engineering degree now-a-days. I’m able to talk the talk enough that they’ve never checked or asked for a transcript.
But it sounds more like your job wanted to, on paper, be compliant with workers rights stuff
“We offer a break at X and Y hours.”
But had a cultural expectation to not follow it. Which is dumb and they can, in my humble opinion, get fucked. Nursing has a massive burnout rate and shit like that is why.
I think you should recognize you dodged a bullet more than you should think about “lying” in an interview.
- Comment on 248 Legally Deceased "Patients" are In These Dewars Awaiting Future Revival - Cryonics 1 week ago:
I gotta see my spell work damn it!
- Comment on microsoft 1 week ago:
We had users at work who swore by that so hard that my boss bought licenses for “Classic Shell” that just did that.
Usefulness wise, the pinned program and search method is objectively much better.
- Comment on microsoft 1 week ago:
I’m gonna say it.
I thought windows 8/8.1 were very fun
- Comment on From Snoop Dogg to Lap Dogg 1 week ago:
That really adds to the evidence that he’s always been a sell out.
Its obvious in retrospect but now that he’s a right wing grifter its hard to avoid now.
- Comment on Best Practice Ideas 1 week ago:
That’s what I was thinking too. Ijust feel better having another layer between the open web an my server
- Comment on Best Practice Ideas 1 week ago:
True but I’m wanting to run VMs for specific tasks as well (mostly game servers) and I like to virtualize Linux distros I wanna play with as well as keep a running windows VM for the off chance I need windows ever.
- Comment on Best Practice Ideas 1 week ago:
Apartment is too small and my partner is too noise sensitive to get away with a rack. So my localLLM and Jellyfin encoder plus my NAS exists like this this summer. Temps have been very good once the panels came off.
- Comment on Best Practice Ideas 1 week ago:
Ansible is next on my list of things to learn.
I don’t think I’ll need to dedicate all of my compute space to K8s probably just half for now.
- Comment on Best Practice Ideas 1 week ago:
The rats nest is behind itI need to re do some of the wiring.
I have all 4 power cables braided and zip tied together with the single data cable so its nice to pull out and put back into the entertainment center.
Only problem is I only had four 1 foot Ethernet cables and three 7 foot cables. So I used the 1 footers for the Pis and the 7 footers are bundled up as best I could and neatly hidden.
I’m waiting on some color coordinated .5 foot cables from Cables and Kits and I will swap the switch and patch panel. I want the Pis to have that one cable and that’s it, but I also want all the patch panel ports to work.
- Comment on Best Practice Ideas 1 week ago:
Rack shown
- Comment on Best Practice Ideas 1 week ago:
This is a 52pi 10 inch rack 8U
I have 4 raspberry Pi4’s 4gb running with POE
Some TP-link gigabit switch with 4/poe ports
3 Thinkcentre Tiny with a ryzen 5 2400GE 32 gigs ddr4 RAM, 512 Sabrent PCIE Gem 4 NVME boot/VM drive, 512 PNY Sata SSDs for databases
I have a bigger server for AI stuff and storage. This is just Tue “production” server for my websites and Git repos
I stole the set up idea from my man Jan Wildeboer