Roark66

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 people protested in the millions.

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.

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 should go to jail if convicted.

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 they do care.

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 GLM5.2 paying $4 per mln tokens I would be burning at least $20-$30 a day.

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 replaces firebase MCP spec with my own that tells the model to use crawl4ai and SearXNG instead.

I've been using this wia Librechat with every commercial and open weight model I tested.

The search is way better than OpenAI and what ClaudeCode uses, but Gemini is way faster. That will change soon as I'm planning to put these instances in a DC with gigabit pipe.

Firebase is not cheap, but it retrieves everything, bypasses captchas and so on.... As long as one uses it for 1% of Web queries the cost is manageable.

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 Austrian painter in my primary school library for example.

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 and are fully quiet.

I have over a meter of radiators there.

Fun fact, I bought my first rtx3090 4 years ago. A year ago I bought another one and they are still the same price used.

I may buy another one (for my servers)

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 to deploy remotely via ipmi/pxe boot.

Then see how it makes stupid mistakes along the way.

Today's AI is pretty amazing when it comes to fixing narrow problems (or creating Web apps with no infra). Give it anything where it needs to go online, download some helm templates and look through them to figure out parameters, as well as write an app and it will make lots of mistakes in seemingly simple stuff.

Opus seems to be the model that works the best with this.

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 these small models got we are still not "at breakeven" for me.

What is "breakeven" with LLMs? For me it is when I no longer have to read the actual code it wrote. I can trust that if I told it to implement and document a certain architecture it actually did that with no stupid mistakes.

The first model ever that did that for me was the first opus. 4.4 if I remember correctly.

The second model was Gemini 3 Pro preview. For few weeks. Then it was lobotomised. I guess it was too expensive to run and they quantized it too hell.

Only Opus remains. If this GLM model truly rivals even an old opus I'll be very happy when day comes that I'll be able to run it locally.

I actually believe you... I was sure I can taste certain spicy things on their way out if you know what I mean :-D

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.

On dual rtx3090 it runs at 140tok/s with a short prompt... Not bad.

Qwen 3.6 dense runs at 40tok/s

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. There is no need to overcomplicated. - ability to manage prompts per model, compress context using alternate models, and minimise token costs better - like the YouTube's Sentdex's Minion mini harness (in fact I'm building on top of his as we speak). - support for agent work fanout - support for MCP, but switchable off/on depending if needed (I use a single MCP aggregator anyway so mcp tool use doesn't eat my context) - support for lsp/tree-sitter, again switchable when needed. - support for OpenAI api and written easily enough so other ones like deepinfra are easy to add.

Nice to have: - have some sort "prompt library" that would store tweaked versions of prompts for different models so it adjusted the harness as needed depending on which model we call.

That's it.

2bits is a joke for serious work. You'd be better with Qwen3.6 under 30G probably.

But there are EU only providers for GLM5.2. For example tensorx. Depending on your definition of "secure" it may be acceptable.

Has anyone compared the costs between maxing out a Claude Max x5 subscription (one for €120 euro a month) and same amount of work on GLM5.2 via API at a cost of $4 per mln token out?

I have a feeling Anthropic may still come out cheeper (mainly thanks to enterprises subsidising the Max subscriptions).

But I'm very excited with the possibility of using fully EU based inference rivalling Opus in quality.

It was surprising to see incidence of death by cesarean is almost 13 per 100k. It is commonly thought as the safest way and half of all births in my country are via cesarean.

They will be hiring frantically next year.

These companies seem to be led by mind bogglingly short sighted people. Maybe they should be replaced by AI. It can't possibly get any worse than now, right?

Did you know human overies contain neurons? I suppose memories are not stored there :-) but still the fact is rather surprising.

Eventually, when all your savings are burned through, when you might need to sell your house, you really wonder if that's what they wanted and if all that was worth it.

I will never accept a country where things like this happen routinely as civilised.

There are ways to implement this without affecting anyone's privacy. For example here in Poland everyone has an ID and everyone that wants to use egov services, file taxes, or check their prescriptions online has a cryptographic identity.

It would be trivial to implement a challenge that would work like this: - the site requests you to sign a challenge

- you sign the challenge and provide it to third party verifier. (can be a gov site or a private company)

- service verifies your signature and gives you the copy of the challenge signed by itself only so your gov ID is never revealed.

- you supply the signed challenge to the original site. You prove you're an adult.

Being Polish and knowing our politics let me tell you how this will go down. The current gov is the one that will swallow everything that comes from the EU (because they are led by a guy that was the deputy leader of the biggest EU parliament party). Same guys that wasted billions trying to implement the e-prescription service that was implemented by the next gov in a fraction of time and cost.

In short they are horribly incompetent on tech so I expect the first version of this will be as bad as the crypto laws they proposed.

But thankfully the president is not from their camp and he will not sign any BS (if it starts looking like he might there will be mass protests - just look up the scale of protests last time they tried to censor the internet)

And then they will be forced to change it.

There is no technical reason why you should disclose your identity to proove you have one. That is the entire point of having those crypto identities. That you can manage how much information you provide.

Here in Poland people will start caring when this verification will be anything other that charge your (debit) card some amount.

No one will accept using their official electronic ID with Facebook etc

If that is how they implement it I expect a massive exodus from these platforms.

No. I've tried all the OS models up to Qwen 480B and Kimi (the biggest models). None come even close to Claude.

I do mostly scripting, devops, data processing and systems stuff (ansible playbooks, managing network devices, deploying new software for various things that involves reading docs, writing helm charts, modifying existing ones etc).

All other models Gemini, Chatgpt, grok and all OS models don't come even close. I'd rather use Sonet than Qwen.

It's a sad reality. I was thinking about implementing maybe some sort of "sanity checking" by running every prompt twice on two different models doing sanity checking of the first on the second.

Elaborate knowledge systems help a little, but personally I think Anthropic must be doing something "clever" with its models (processing via multiple models etc). Nothing else in my mind explains the discrepancy.

This is so funny, especially that in the current "Big Co" I'm working at we get constant pressure on "Every team must use agents" for no reason at all despite repeatedly telling the "decision makers" many of us have been using these tools for YEARS and NONE of them can work on actual mature code for more than half an hour let alone a weekend without human in a tight loop.

I improved a similar issue by writing custom instructions for copilot that give it enough context to do PR reviews that are only 30% BS.

I asked other team members to run my custom instructions to perform a review with copilot before they submit...

Of course no one is doing it. It looks like the PRs I get are still straight from copilot. So I tend to run my review prompt. Cut out the 30% BS issues it "finds" and the rest is good.

This is a sign of things to come. First they sabotage your perfectly legal ML dicking around in your homelab.

Next they will be sabotaging anything that competes with them. Oh you are working on OpenCode codebase? Sorry Dave I can't allow you to do that.

How is this not illegal monopolistic practice? It is as if a maker of metalworking equipment put in the ToS you're not allowed to make your own spare parts using said equipment. Those fuckers should be banned from the EU and alternatives should get public funding.

(don't even tell me about these companies being a result of "free market". It is state level oligarchy it's clear to everyone. I don't see why we shouldn't counter them with public funding ourselves).

Just like Taiwan managed to take over advanced semiconductor production a well governed narrowly targeted state level funding will always win with oligarchs trying to do the same (they will always try to skim more and more). Of course I'm talking about things that require many dozens of billions in investment. Far too much for the free market to handle.

Nordstream was privately funded, half by Gazprom, half by a consortium of privately owned large utilities

And do you also believe what Merkel said that "it's purely a commercial venture"?

Who chairs those "privately owned large utilities"?

What are the links between those people and "the establishment" that includes people like Merkel (earlier Shroeder - his work for Gazprom was also purely commercial venture of a private citizen, right?).

That establishment deciding its a great opportunity for Germany to be the Russian gas station of rest of Europe forced to use that gas as the only "green transition" hydrocarbon.

And not only was it a great commercial venture (that had in its profitability calculation getting rid of nuclear - including blocking countries like Poland from building it, squeezing out other countries that own pipelines again such as Poland/Ukraine/Hungary and so on and forced "transition" to gas for the EU - let's not kid ourselves, renewables will never be more than 50% of base load unless battery tech gets cheaper, so that "transition fuel" would last for 50+ years.

It also contained a humanitarian element of giving Putin huge amount of money therefore making sure the dictator will absolutely not use it to build armies to invade his neighbours (despite doing that already at the time in Georgia for example), but he will get used to that money so much he will spend it all and will not want to stop it coming therefore granting eternal peace in Europe.

Anyone who thought the public will swallow this must have been high... But the Germans did.

"nothing to see here" - right?

I for one am glad hopefully the German public realises what kind of state Russia is now, and what "doing business" with them leads to (it corrupts your own country) , but how long that knowledge remains, and why it took a full scale war in Europe to acquire it I don't know.

And that is why I'll not be buying a vw ever again despite being a fan of the brand so far.

I've found empirically calling various models "a stupid c*nt" and berating them otherwise consistently produces better output. Mainly in response to genuine errors.

Although OpenAI and google models are much more responsive to it. With Anthropic if you treat Opus too harshly it might start pushing back if the insults are not justified.

So I'm not surprised they had good results with chatgpt.

A very good reason. I refuse to use paper straws. They are disgusting (it feels like sucking on a piece of printer paper).

Don't get me wrong. I try to remove plastics in all areas of my life as well (because of microplastics), but can't they coat the surface in some biodegradable polymer like PHA/PBS?

Or if this is too expensive coat it with some beeswax at the very least...