npodbielski

6 ms·

61.
by npodbielski 4mo ago
Maybe they have DeepSeek subscription for that occasion?
62.
by npodbielski 4mo ago
Me?
63.
by npodbielski 4mo ago
I am using Mistral Vibe.
64.
by npodbielski 4mo ago
480GB? Probably like 100k$ each? :D
65.
by npodbielski 4mo ago
Yep, I left one project because I think I was underpaid there. I could not find another job for like 3 months. I finally found one and I was also very stressful. No it is better, at the same project and I think I make more money than I woul
66.
by npodbielski 4mo ago
With manual ROM in switches probably not.
67.
by npodbielski 4mo ago
You cant accept freaking EULA without linking your own network to some cloud in the US. So no. No.
68.
by npodbielski 4mo ago
I am running nginx proxy on my public IP on top of ubiquiti UNVR it works via headscale. But honestly I am super happy with their hardware and software but necessity to link your network to their cloud just be able to use 100% of features o
69.
by npodbielski 4mo ago
I would say that it is now very easy to steal 'AI' providers credentials this way. And then you can use them to write more malware or scam or use models for generating speech to call people and get them to 'redeem'. Or a
70.
by npodbielski 4mo ago
Just out of curiosity, where did you move?
71.
by npodbielski 4mo ago
AI is different because you can't encrypt it. An running on someone else's hardware is basically just 'trust me bro! I won't read it!'. Of course you can say that about I.e. database too, but at least you can run it
72.
by npodbielski 4mo ago
I bought r9700 for about 1700-1800$ and I have like 800t/s prompt and about 50t/s of inference on average? It hurt a bit when you change a prompt so llama.cpp have to discard entire cache and it have to think for 2-5min depending
73.
by npodbielski 4mo ago
Exactly what I do! You can have web UI for PI!
74.
by npodbielski 4mo ago
Well there are in a business of selling you quote ubquote, intelligence, so it does what they want it to do: imply intelligence, which for me is just an illusion. But on the other hand this is what politicians and other conmen are doing? Us
75.
by npodbielski 4mo ago
Anybody tried it on Strix Halo?
76.
by npodbielski 4mo ago
I would watch that.
77.
by npodbielski 4mo ago
I think it is harder to remove those from your own speech. I have been doing that for few months now and I still get back at it when I am in hurry or stressed.
78.
by npodbielski 4mo ago
I watch american politics with great interests to see if their will overthrow their own tyrant. That would be interesting though I am not sure if good for the rest of the world.
79.
by npodbielski 4mo ago
Can't you change the domain? If you want to work within any project, enterprise or open source, you have to obey their rules. If you do not like to do that, you do not have to work with them. > Think about that. I thought and I do n
80.
by npodbielski 4mo ago
What is the point of asking LLM to do manual testing? IMHO it would be much better to make it write automated tests. So you can just rerun them?
81.
by npodbielski 4mo ago
I was talking more about middle management. Some smaller CEOs like the parent comment suggested. Those people, as you said, are not our overloads, because they lack intelligence and technical skills.
82.
by npodbielski 4mo ago
For the long time this worked oh right: get to know right people, wipe some asses, lick some other, play some golf and be sociopath. But right now it does not cut it anymore. You have to be either smart, skilled or know your business and IT
83.
by npodbielski 5mo ago
> You can use the f=json search query parameter to get JSON-formatted search results, as per the FAQ entry above. Nice so EU-based company with API available for my agent to use. Right now I am using lightpanda but this my be simpler. I
84.
by npodbielski 5mo ago
Exactly what I thought! And then I read that some guy is working while driving. Jesus Christ... Next time it will be post: I am talking to my phone about my code while I shower and I never felt more productive!
85.
by npodbielski 5mo ago
They know it will tank and want you money to save their friends.
86.
by npodbielski 5mo ago
Ud_Q4_k_xl
87.
by npodbielski 5mo ago
Yes, Qwen 3.6 MoE is hitting like 80-90tk/s on Strix halo. On R9700 I had like 170t/s. It was not possible to keep up. But MoE is circling very often. I switch then to dense model and have 20-30t/s but it is able to solve qui
88.
by npodbielski 5mo ago
Yeah right, it could not. US would probably would be unable to produce something really simple like electrical cables or plastic flip flops, yet alone computer chips.
89.
by npodbielski 5mo ago
sure mate I was just asking you what is your take about that. I mean I am working remotely and barely see my friends, alienate myself even more because I want to listen to private music does not seem like a good idea. Concerts and music fes
90.
by npodbielski 5mo ago
I would just write 'do this'

More ›