proxysna

7 ms·

1.
by proxysna 6d ago
He didn't really want to make MGS4 and MGS5 and then still made successful games. Death Stranding franchise is one of the "games that he is well known for" and it is most def not 20+ years old. Dont see your point.
2.
by proxysna 6d ago
I mean every time that Kojima made a game it turned out to be something that people talk about for years on. But the second it gets slightly more ambitious investors get their panties in the knot about it even though throughout history Koj
3.
by proxysna 14d ago
Utterly awful article. R1? Llama3.1? Not being able to serve larger llms on 8(!) RTX pro’s? You can literally run open weight SOTA models with relative ease. Even 4 GPUs get you there with a bit of elbow grease and compression. Pure slop.
4.
by proxysna 1mo ago
My brother-in-law had this daily for 6 months while being in custody. Now he cant even stand near food carts that serve it. Never really liked it myself, because if there is a choice available, there is most likely a better dish right next
5.
by proxysna 1mo ago
Really hoped to see their hw out in the wild one day
6.
by proxysna 1mo ago
You can run it on a anything. Now officially. About a year back i had proxmox cluster on 3 orange pi 5+. As long as you have a some sort of Debian it's pretty seamless.
7.
by proxysna 2mo ago
> My primary concern is the risk that authoritarian governments—not solely the Chinese Communist Party (CCP), although the CCP is clearly the most capable threat—build AI models that are more powerful than those built by the US, Begging,
8.
by proxysna 2mo ago
Never used HyperCard, but from descriptions and screenshots in the article reminds me a lot of Delphi/Pascal/Lazarus IDE’s. Appeal is understandable, because it is so cool to be snore to from a few lines of code to a clickable int
9.
by proxysna 3mo ago
Deploying and building OpenTTD if my favourite exercise when i am trying to learn new CI tool or orchestrator or packaging format. Simple enough, yet fun.
10.
by proxysna 3mo ago
Nice. Thanks for the insight!
11.
by proxysna 3mo ago
I've been using deepseek for some development at home and it is really good for the price. It is at the point where i am ok with using it as a tool that i can rely on and not an expensive gadget with flaky uptimes.
12.
by proxysna 3mo ago
I'd would assume that this is not a monolithic cluster of 40k vm's but at least tens of clusters. Which puts it in the realm of capabilities of Proxmox.
13.
by proxysna 3mo ago
Great time to migrate off VMware. All the migration paths are well-trodden by now, but goddamn 40k vm's. A lot of work ahead.
14.
by proxysna 3mo ago
Alpine is my go-to nowadays for everything in my homelab except desktop (I use Void btw), because of how dirty the setup to make GPU's work with musl kernel.
15.
by proxysna 3mo ago
Looks really nice, but 10 fps in Firefox.
16.
by proxysna 3mo ago
It's just a DGX spark with faster memory and a windows boot?
17.
by proxysna 4mo ago
There is a docker registry under the same name. https://zotregistry.dev
18.
by proxysna 4mo ago
I think hashicorp still have an implementation for vaults seal/unseal process. Unless something changed ofc
19.
by proxysna 4mo ago
Feels about right. I've launched an internal demo of Claude Code and Deepseek on the same day and we burned through our monthly allowance for Claude in just over a week, with more than a half of that budget being spent in one day. With
20.
by proxysna 4mo ago
“Expert” that does not know what a Terraform is. lol, lmao even
21.
by proxysna 5mo ago
Cool, i am mostly a plumber for these things, but do you have any sort of reading that i can go through to wrap my head around it to some degree?
22.
by proxysna 5mo ago
I've used only Qwen3.5 so far for work and was, after initial struggles, successful with GPU setup, no mlx. Ngl the fact that they are using `presence_penalty: 0` and no `max_tokens` is weird after that exact setup caused me "ini
23.
by proxysna 5mo ago
You need to set sampling parameters for the llm. Had the same issue with Qwen3.5 when i first started. You can grab them off the model card page usually. From Qwen3.6 page: Thinking mode for general tasks: temperature=1.0, top_p=0.95, top_k
24.
by proxysna 5mo ago
OpenWebUI with postgres and vllm for inference, searxng for websearch a few other mcp's for tools.
25.
by proxysna 5mo ago
This is from an example from my Nomad cluster with two a5000's, which is a bit different what i have at work, but it will mostly apply to most modern 24G vram nvidia gpu. "--tensor-parallel-size", "2" - spread the L
26.
by proxysna 5mo ago
We do make Claude and Mistral available to our developers too. But, like you said, security. I, personally, do not understand how people in tech, put any amount of trust in businesses that are working in such a cutthroat and corrupt environ
27.
by proxysna 5mo ago
Qwen3.5-27B with a 4bit quant can be run on a 24G card with no problem. With 2 Nvidia L4 cards and some additional vllm flags, i am serving 10 developers at 20-25tok/sek, off-peak is around 40tok/sek. Developers are ok with that p
28.
by proxysna 5mo ago
Yeah, my thought exactly, execution lacked, but i do admire the attempt.
29.
by proxysna 5mo ago
What a great write up, and a video too! Even though Minecraft stuff ofc was a bit of a bait, it would be interesting see the answer to "Can it run Doom?".
30.
by proxysna 5mo ago
What a load of e-waste. Nightmare.

More ›