Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
Roark66
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
4 ms
·
31.
▲
by
Roark66
1mo ago
>> I've never been the kind of coder >Looks like you were never really a coder, to be honest. I do not understand the bitterness here at all. AI is just a tool you can use for better or worse. I learnt basic programming at an
32.
▲
by
Roark66
1mo ago
Soo, what are they buying exactly? Can I setup a forgejo instance, upload a bunch of open source models to it and get bought out for $13bln? I fail to see the point of this. Their inference hosting is a joke (I don't think I ever had a
33.
▲
by
Roark66
1mo ago
Unsloth doesn't have all quant versions yet :-(
34.
▲
by
Roark66
2mo ago
The page says 170G/s memory bandwidth for the NPU and 1.2T/s for the GPU. Why the discrepancy if it's all "unified memory"? The former is nothing to write home about as far as AI compute is. The latter is really nic
35.
▲
by
Roark66
2mo ago
You know what. As a big Anthropic fan that pays €120 a month for Claude Max x5 and maybe €30 a month for API access I got extremely annoyed at the extremely variable quality of service I'm getting. It is not only that every single turn
36.
▲
by
Roark66
2mo ago
The problem is benchmarking. Not everyone has a 500k token workstream of the model they are setting up for the first time to run it against 10 different config and compare differences. And if you download benchmarks from the net they are li
37.
▲
by
Roark66
2mo ago
A somewhat similar situation here in Poland. While the volumes of these messages are much, much lower (the most i ever got was one or two a week. Normally there is maybe one every month or two) they are typically about bad weather. Somethin
38.
▲
by
Roark66
2mo ago
I've been using Qwen3.6 models locally for a couple of weeks. Both the A3B moe and the dense variant. The moe works well in Librechat combined with my local search/Web retrieval system. All components use open source projects such
39.
▲
by
Roark66
2mo ago
It is worth mentioning DGX is about a third up to a half performance of a 6 year GPU Rtx3090... I prefer to stay with my 3090s.
40.
▲
by
Roark66
2mo ago
How about implementing proper permissions on the tool use or if you need more flexibility a dedicated model to analyse potential impact? (Like Claude Code's Autopilot but more configurable)? I find solutions like this to be a like tryi
41.
▲
by
Roark66
2mo ago
It is quite funny an EU company patenting a software feature that is basically unpatentable in EU in the US. Clearly this is an attempt to prevent similar patents from being weaponised against them in the US. No one cares about such stuff i
42.
▲
by
Roark66
2mo ago
This will not fly anywhere outside France Polish here and the very first question I have is "who would decide which cultural industry representatives would get the money"? And what right the decision makers have to decide that. Ho
43.
▲
by
Roark66
2mo ago
You know there is such a thing as law, "EU" can't just block a merger or take away their profits beyond a certain fee. The fee is the only enforcement mechanism in the law. If you decide it is worth for you as a company to pa
44.
▲
by
Roark66
2mo ago
Open code is cool once I added the ctrl-o function to it (show thinking and command outputs at will instead of on by default), but sadly I don't think it got merged by the team. I forgot why. But I can't use an AI cli without that
45.
▲
by
Roark66
2mo ago
This is very cool that they open weight it. Are there any EU providers serving it yet?
46.
▲
by
Roark66
2mo ago
That is a good analogy. But the limitations of LLMs seem to be where they lack some information they tend to hallucinate/invent. The rare benchmarks that measure "knowing what the model knows it doesn't know" show us the
47.
▲
by
Roark66
3mo ago
Exactly, also the ECJ is horribly corrupt (search for the corruption scandal revealed by a French journalist few years ago). Nothing happened back them. This will never become law in my country(Poland). Last time they tried to push it peopl
48.
▲
by
Roark66
3mo ago
What? 75% people use personal chats for work? I've never seen this in my over 20 years working in the UK, Poland and few other countries. Personal messaging apps were always banned everywhere I worked.
49.
▲
by
Roark66
3mo ago
A good example. In cases where social media are clearly intentionally harmful they should be prosecuted under existing laws. Instead of blocking under 16 year old from social media we should fix it. Via courts if needed. Also the CEOs shoul
50.
▲
by
Roark66
3mo ago
Recently I started getting messages from Clause Code (on a plan). "You're restoring an old session are you sure you don't want to compress the context? This will use a substantial amount of your usage quota" So it seems
51.
▲
by
Roark66
3mo ago
How is GLM5.2 cheaper for you then Opus? I assume you're an individual so you get Max pricing like me, not API. I have Max x5 for 120Eur a month. I use it a lot (but usually I don't multitask). I almost never hit the limits. With
52.
▲
by
Roark66
3mo ago
The author mentions lack of good Web search. I've been using slightly modified crawl4ai and searXNG together with firebase for the rare sites that insist on throwing wrenches in the works of my LLMs. I also have my fork of metamcp that
53.
▲
by
Roark66
3mo ago
There are no books banned in EU... Some countries have laws that criminalise glorification of nazism or communism, but I never heard any book was "banned" as a result. Here in Poland we had "Mein Kampf" by certain Austri
54.
▲
by
Roark66
3mo ago
I have to see I'm in the rtx camp. A dual rtx3090 workstation with 200G of ram and zen5 9950x cpu. All watercooled. The only reason I can tell it's on, is the very quiet hum of the slow speed water pump. Large fans run at 1200rpm
55.
▲
by
Roark66
3mo ago
Serious programming is dealing with a large knowledge surface area. So not "implement me a shading algorithm" But more like: make an multi user app running on a k8 cluster, design the whole thing to be indempotent, scalable, easy
56.
▲
by
Roark66
3mo ago
I think there is no reasonably priced machine you could run locally to do serious work with LLMs... 10x rtx6000 Pro in a large workstation is probably the way to go for someone wanting to run GLM5.2. Other than that it is cloud. As good as
57.
▲
by
Roark66
3mo ago
I actually believe you... I was sure I can taste certain spicy things on their way out if you know what I mean :-D
58.
▲
by
Roark66
3mo ago
Poland. Both children my sister had were cesarean. There were no complications at all, but both were as the doc put it "large" so my sister was given a choice.
59.
▲
by
Roark66
3mo ago
On dual rtx3090 it runs at 140tok/s with a short prompt... Not bad. Qwen 3.6 dense runs at 40tok/s
60.
▲
by
Roark66
3mo ago
If we programmers united we had a clouded code alternative that didn't suck :-) But we're not far. My requirements: - a terminal app without advanced tui, not written like "a browser running in a terminal" or a game. The
More ›