Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Coding with LLMs in the summer of 2025 – an update

antirez.com

Post reply on HN
Print view View original

391–400 of 434 posts

Re: Coding with LLMs in the summer of 2025 – an update

#391

Post by cutemonster »

rapind wrote: post #319
worldsayshi wrote: post #281

Earlier quoted context omitted.

> If the gains don't plateau, well then we're obsolete anyways I think there's room for more nuance here. It could also be a situation of diminishing returns but not a sharp plateau. That could favour the big players. I think I find that scenario most likely, at least in between major breakthroughs.

Well diminishing returns will have the same effect as a plateau. If you're on a log with your (much cheaper, Chinese) competition, then your advantage is very quickly microscopic.

Can't AIs plateau at a prohibitively hight cost, so only the biggest companies can build the really good ones.

Search engine tech isn't that much of a secret nowadays? Still it's prohibitively expensive for almost everyone to build a competitive search engine. What if really good AI turns out to be more like that (both training and inference)

Re: Coding with LLMs in the summer of 2025 – an update

#392

Post by wizee »

KronisLV wrote: post #366
dakiol wrote: post #25

> Gemini 2.5 PRO | Claude Opus 4 Whether it's vibe coding, agentic coding, or copy pasting from the web interface to your editor, it's still sad to see the normalization of private (i.e., paid) LLM models. I like the progress that LLMs introduce and I see them as a powerful tool, but I cannot understand how programmers (whether complete nobodies or popular figures) dont mind adding a strong dependency on a third part…

The software is largely there: you can run Ollama, vLLM or whatever else you please today. The models are somewhat getting there: even the smaller ones like Qwen3-30B-A3B and Devstral-23B are okay for some use cases and can run decently fast. They’re not amazing, but better than much larger models a year or two ago. The hardware is absolutely not there: most development laptops will be too weak to run a bunch of tool…

On my M4 Max MacBook Pro, with MLX, I get around 70-100 tokens/sec for Qwen 3 30B-A3B (depending on context size), and around 40-50 tokens/sec for Qwen 3 14B. Of course they’re not as good as the latest big models (open or closed), but they’re still pretty decent for STEM tasks, and reasonably fast for me.

I have 128 GB RAM on my laptop, and regularly run multiple multiple VMs and several heavy applications and many browser tabs alongside LLMs like Qwen 3 30B-A3B.

Of course there’s room for hardware to get better, but the Apple M4 Max is a pretty good platform running local LLMs performantly on a laptop.

Re: Coding with LLMs in the summer of 2025 – an update

#393

Post by HarHarVeryFunny »

edanm wrote: post #348
dakiol wrote: post #25

> Gemini 2.5 PRO | Claude Opus 4 Whether it's vibe coding, agentic coding, or copy pasting from the web interface to your editor, it's still sad to see the normalization of private (i.e., paid) LLM models. I like the progress that LLMs introduce and I see them as a powerful tool, but I cannot understand how programmers (whether complete nobodies or popular figures) dont mind adding a strong dependency on a third part…

> The excuse "but you earn six figures, what' $200/month to you?" doesn't really capture the issue here. Why? If I want to pick up many hobbies, not to mention lines of professional work, I have to pay for tools. Why is programming any different? Why should it be? Complaining that tools that improve your life cost money is... weird, IMO. What's the alternative? A world in which people gift you life-and-work-improving…

> Btw, I think this was actually less true in the past. Compilers used to cost money in the 70s/80s. I think it's actually cyclical - most likely tools will cost money today, but then down the line, things will start getting cheaper again until they're free.

I don't see it as an inevitable cycle. Free tools (gcc, emacs) and OS (Linux) came about as idealistic efforts driven by hobbyists, with no profit goal/expectation, then improved as large companies moved to support them out self-interest. Companies like RedHat have then managed to make a business model out of selling support for free software.

Free LLMs or other AI-based tools are only going to happen where there are similar altruistic and/or commercial interests at play, and of course the dynamics are very different given the massive development costs of LLMs. It's not a given that SOTA free tools will emerge unless it is the interest of some deep-pocketed entity to make that happen. Perhaps Meta will develop and release SOTA models, but then someone would have to host them which is also expensive. What would the incentive be for someone to host them for free or at-cost?

Re: Coding with LLMs in the summer of 2025 – an update

#394

Post by irthomasthomas »

dakiol wrote: post #25

> Gemini 2.5 PRO | Claude Opus 4 Whether it's vibe coding, agentic coding, or copy pasting from the web interface to your editor, it's still sad to see the normalization of private (i.e., paid) LLM models. I like the progress that LLMs introduce and I see them as a powerful tool, but I cannot understand how programmers (whether complete nobodies or popular figures) dont mind adding a strong dependency on a third part…

Kimi K2 competes with these. Even beats them in some evals.

Re: Coding with LLMs in the summer of 2025 – an update

#395

Post by headcanon »

> Gemini 2.5 PRO | Claude Opus 4

Glad to see my experience is reflected elsewhere. I've found Gemini 2.5 PRO to be the best bang-for-buck model: good reasoning and really cheap to run (counts as 1 request in cursor, where opus can blow my quotas out of the water). Code style works well for me too, its "basic" but thats what I want. If I have only one model to take to my deserted island this is the one I'd use right now.

For the heady stuff, I usually use o3 (but only to debug, its coding style is a bit weird for me), saving Opus 4 for when I need the "big guns".

I don't have Claude Code (cursor user for now), but if I did I'd probably use Opus more.

Re: Coding with LLMs in the summer of 2025 – an update

#396

Post by rapind »

cutemonster wrote: post #391
rapind wrote: post #319

Earlier quoted context omitted.

Well diminishing returns will have the same effect as a plateau. If you're on a log with your (much cheaper, Chinese) competition, then your advantage is very quickly microscopic.

Can't AIs plateau at a prohibitively hight cost, so only the biggest companies can build the really good ones. Search engine tech isn't that much of a secret nowadays? Still it's prohibitively expensive for almost everyone to build a competitive search engine. What if really good AI turns out to be more like that (both training and inference)

The issue with launching a search engine company is probably mind share more than anything else. Once Google was a verb, it was pretty locked in. Even so, there are alternatives that some people use and find superior, like Kagi and DDG. Now you're seeing a lot of people who just use ChatGPT instead of google for their searches.

For AI, I think that ship has sailed already. OpenAI is the closest to dominance, but not currently the best at all tasks (claude and gemini for some tasks), and everyone else is nipping at their heels, followed by open / cheap models anywhere from 6 months to 1 year behind. Maybe I'm wrong, but as far as I can tell, all evidence points to it becoming a commodity (or utility), similar to cloud computing.

For example AWS is pretty dominant in the cloud computing space, but the differences aren't really that big of a deal for most people and there are services that will extract the cloud for you as a generic services layer. Like OpenRouter does for AI models. Is AWS really better than other cloud providers? Maybe, in some situations, with some requirements, but it's definitely not the general rule. Their focus since the beginning has been layering value on top of this base cloud offering (I'd argue to the point where there are so many services it's confusing AF), and I think it's the same with AI providers. It's even the same players for a lot of it (Google Cloud, AWS, Azure all offer AI services).

Re: Coding with LLMs in the summer of 2025 – an update

#397

Post by jmb99 »

pmarreck wrote: post #316
jmb99 wrote: post #313

Earlier quoted context omitted.

What’s your budget and speed requirement? A quad-CPU Xeon E7 v4 server (Supermicro X10QBI, for example) with 1TB of RAM gives you ~340GB/s memory bandwidth and enough actual memory to host a full DeepSeek instance, but it will be relatively slow (a few tokens/s max in my experience). Up front cost a bit under $1k, less if you can source cheap 32GB DDR3 RAM. Power consumption is relatively high, ~1kW under load. But I…

This would require massively more power than the Mac Studios.

Yep, ~1kW as mentioned. Depending on your electrical rate, break even might be years down the line. And obviously the Mac Studios would perform substantially better.

Edit: And also, to get even half as much memory, you need to spend $10k. If you want to host actually-large LLMs (not quantized/distilled versions), you'll need to spend close to that much. Maybe you can get away with 256GB for now, but that won't even host full Deepseek now (and I don't know if 512GB either, with OS/etc overhead, and a large context window).

Re: Coding with LLMs in the summer of 2025 – an update

#398

Post by freedomben »

rolisz wrote: post #325
freedomben wrote: post #225

Earlier quoted context omitted.

Indeed. I've been a vim user for almost two decades, and it's been a long, long time since I"ve had to spend time solving problems/optimizing my tooling. Yes it was a big up front investment, but it's paid off immensely. I don't think I'm anything special so please don't misunderstand this as a brag, but I routinely have people enjoy "watching" me use vim because I can fly around the codebase with lightning speed, of…

What kind of crappy machines do your coworkers have? VS code takes 2-3 seconds at most for me to load.

They mostly have macbooks (I think either M1 or M2?). I don't know if they are loaded with extensions or something that adds time to it, but it definitely takes longer than 2-3 seconds

Re: Coding with LLMs in the summer of 2025 – an update

#399

Post by amelius »

loudmax wrote: post #388
amelius wrote: post #376

Earlier quoted context omitted.

Didn't Karpathy, in his latest talk, say something along the lines of: don't bother with less capable models, they are just a waste of time.

It probably depends what your objective is. One of the benefits you get from running less capable models is that it's easier to understand what their limitations are. The shortcomings of more powerful models are harder to see and understand, because the models themselves are so much more capable. If you have no interest in the inner workings of LLMs and you just want the machine to spit out some end result while putt…

Is it really possible to learn from the mistakes of an LLM? It sounds like psychology or even alchemy 2.0, to be honest.

Re: Coding with LLMs in the summer of 2025 – an update

#400

Post by skydhash »

LeafItAlone wrote: post #373
skydhash wrote: post #209

Earlier quoted context omitted.

Emacs is not a python oriented IDE. So the comparison is moot from the beginning. Some people likes what Emacs offers and mesh it with external tools. Some prefers a more complete package. Nothing wrong with either, especially if you have the time and the knowledge to iterate quicly on the first. What you may need is something others can do without. So what’s best is always subjective.

Sure. I think it’s odd for you to criticize someone’s reasons when you apparently just reject their underlying premise to begin with (which is, effectively, my experience with a particular open source tool isn’t the best for me and the one I found that does isn’t open source). Frankly, the idea of having 4 different ways of editing code, as you’ve described above, just seems like a nightmare to me. I like to have one…

The point I was making was against conflating optimizing tooling and wasting time not solving problems. Not about going with an IDE or not.

Everyone works differently. For most tasks, I prefer Emacs. But for pairing, VS code may be more familiar to others. And it’s hard to get rid of Android Studio and XCode if you’re doing mobile development.

No one can judge workflow and tooling choices as long as the results are in.

Post reply on HN