Qwen models are actually very competitive with frontier models, and you can run them on your local computer. Gotta have a decent graphics card and by that time the current cost of the rig may not justify it over paying $100/month for cloud model but it’s all out there.

Qwen is still controlled by Alibaba, one company. We can't let the future be in the hands of a few companies, can we?

Fun fact: Qwen was not initially a Apache Licensed project, it was based on a custom license from Alibaba that restricts commercial use: https://github.com/QwenLM/Qwen/blob/ba2d85a13b28ed1ee0dde2d6.... There's no guarantee that they won't just switch it back later.

Kudos for them for switching to Apache License, of course. BUT, they're still a for-profit company. So as DeepSeek btw.

>Gotta have a decent graphics card and by that time the current cost of the rig may not justify it over paying $100/month for cloud model but it’s all out there.

Never, ever, subscribe. When you subscribe, they win. They cornered the silicon market to force you to subscribe. Don't be a sub, or at least keep your sub tendencies in the bedroom. ;^)

Fluctuating token costs make it worth it

Please don't over-promise. -- An AI open³ dev.

Well, it was great while it lasted - I had fable build me a bunch of stuff this week that opus was just screwing up too much and could never finish. Good thing there are plenty of choices now even if US gov fucks up US AI.

What else are you using. Same experience, chatgpt isn't good enough and Opus is not smart

Pro/max subs are not as flexible as bedrock in api use and don’t seem to run the same models either - often times they are notably dumber (quantized I guess) than bedrock equivalent.

Huh, I felt the inverse.

I’ll ask it to write me some win32 ui crap when I get hands on it, it will need all its brainpower to get that idiocy right.

Opus regularly bitches and wines to me how long something will take and that I should think before asking it to do it. But then it does it anyway in 15 minutes.

considering they work with any architecture/configuration given enough compute, just more or less efficiently - then maybe it's fundamental, in the same sense as why electricity works...

I like gpt oss - great model even if not too smart.. runs on my laptop at over 100ts has a certain tone that I like over all these qwens stuck up their asses.

27b is slow as molasses vs 35b on local stuff I have (m5 max). Mtp doesn’t make any difference either.

…but they reason well enough given enough context (using their matmuls).

To this day frontier models think that A and not B means A and B when the sentence gets pushed far enough back in their context window. The context length that model can reason over without obvious errors is much smaller than the advertised context. Between a 1/4th to a 1/20th what is advertised on the tin.

Critiques like this tend to focus very hard on what models can't do. It's true, they have limitations.

But they're also superhuman in so many other ways. It's valid to point out limitations, but that doesn't support the conclusion that models are not incredibly powerful and capable of the functional equivalent of reasoning at human or superhuman levels in many scenarios.

They may be better than humans at reasoning but they are substantially worse than the first generation logic programs from the 1950s.

These types of comments help demonstrate first-hand how human reasoning stacks up against what an LLM would say in this situation.

Agreed. Both are true. I sometimes think of the calculator as being superhuman as well.

Yes, although the calculator couldn't "reason" the way ML models can.

All the political and emotional reactions to LLMs seem to obscure how absolutely amazing this technology is. I've pointed them at codebases I wrote entirely myself and had them find bugs, point things out I had missed, plan and implement refactorings to improve code quality, etc. I may be "smarter" than the models in some ways but there's no question they're smarter than me in others. They're unlike any tool we've ever had access to.

Yes, the politics and economics around them leaves a lot to be desired (read: is absolutely terrible), and there are a lot of valid justifications for the "AI backlash", but there's a very important baby in that bathwater.

Do you also happen to remember what you ate last thrusday?

"If you have a question look in the specification for the answer and don't just guess" seems a fairly important thing to remember for more than a couple of minutes...

I had a coding session where I was doing stuff across two repositories. And CC forgot in exactly which repository a particular file was so it was grepping the parent directory. I just asked it to write all important key-value pairs which it thinks are important to a file and it never did parent directory grepping.

Do you have a point? Because last time I checked, AIs were supposed to be better than us fragile faulty humans, and weren't designed to emulate us and all our faults.

If you have been following the news, harness is also a scaling direction now. Prompt your AI better not to forget relevant stuff or write them in a file which it can refer later. This way context can be refreshed, this is cached facts method or rolling window method of refreshing your memory just like you would ask a colleague to explain a concept again. These are solved problems.

Are they though? Because I really shouldn't have to use Claude Code (and I don't) just to get even decent results. As I said, I thought one of the biggest advantages AI was supposed to have was that it wouldn't need such constant reminding of things because it wasn't trying to emulate us faulty, forgetful, fragile humans who do have memory loss?

You can convert your best practices into a skill or best practices md file and CC will keep that in purview.

Mist of human written codebases are unusable for llm dev by that definition.

Turns out that if they're unusable by LLMs they're likely unusable by human devs. If you follow sane clean coding principles (like not having godclasses) it turns out coding agents (and humans!) can understand and navigate your codebase, especially if you use recognizable patterns, even with very light documentation.

One of these days you’ll learn about “enterprise” code

Agree, agentic coding seem to have shifted the trade-off about over-engineering, I found clean architecture is a good practice for coding agent, so every task have a clean and limited context, only a few directly connected classes or interfaces is relevent to any local modification.