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

311–320 of 434 posts

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

#311

Post by asadotzler »

ls-a wrote: post #273
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…

Programming on/for Apple has never been free. So it's not a surprise to some engineers. You're right that programming might become the way of Apple in the future. However, I think engineers should rejoice because AI is the best thing that happened to them.

I paid ~$600 for my first Windows compiler, over a grand in today's money. But I didn't have to keep paying every month forever to be able to code at all. Take Claude or whatever away from a vibe coder and they're completely dead in the water.

Apple's fee is like that Visual Studio purchase I made, a fee that lets me compile for their platforms. It's not a subscription without which I can't code anything at all.

Creating a new dependency on monthly subscriptions to unsustainable companies or products is a huge step away from accessible programming of the last 50 years and one that should not so casually be dismissed.

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

#312

Post by asadotzler »

floucky wrote: post #72
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…

Why do you see this as a strong dependency? The beauty of it is that you can change the model whenever you want. You can even just code yourself! This isn't some no-code stuff.

Change the model, learn how to talk to a new and poorly documented model, and get entirely different results. Yep, easy as pie.

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

#313

Post by jmb99 »

simonw wrote: post #141
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 models I can run locally aren't as good yet, and are way more expensive to operate. Once it becomes economical to run a Claude 4 class model locally you'll see a lot more people doing that. The closest you can get right now might be Kimi K2 on a pair of 512GB Mac Studios, at a cost of about $20,000.

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 don’t think you can self host a large model cheaper than that.

(If you need even more memory you could equip one of those servers with 6TB of DDR3 but you’ll lose a bit of bandwidth if you go over 2TB. DDR4 is also a slightly faster option but you’re spending 4x as much for the same capacity.)

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

#314

Post by asadotzler »

paulddraper wrote: post #254
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…

And you thought Visual Studio was free? Or Windows? Or an Apple Developer License? There are free/free-ish, options, but there have always been paid tools.

Paid tools were rarely monthly subscriptions without which you could not produce code or executables.

Further, I can write and compile an application for Mac, Windows, or Linux with entirely free tools including tools provided directly by Apple and Microsoft.

This discussion is about is a monthly subscription, without which, most up and coming coders, certainly "vibe coders," are completely dead in the water.

These two dependencies are not the same. If that's not obvious to you, I don't know what else to say to that.

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

#315

Post by Karrot_Kream »

vl wrote: post #309

I use Claude Code with Opus, and article recommends Gemini 2.5 Pro. I want to try it as well, but I don't know a tool which would make experience compatible to Claude Code. Would it make sense to use with Cursor? Do they try to limit context?

Gemini has its own Claude Code like tool you can use. https://github.com/google-gemini/gemini-cli

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

#316

Post by pmarreck »

jmb99 wrote: post #313
simonw wrote: post #141

Earlier quoted context omitted.

The models I can run locally aren't as good yet, and are way more expensive to operate. Once it becomes economical to run a Claude 4 class model locally you'll see a lot more people doing that. The closest you can get right now might be Kimi K2 on a pair of 512GB Mac Studios, at a cost of about $20,000.

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.

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

#317

Post by andrew_k »

vl wrote: post #308
ghm2180 wrote: post #116

Earlier quoted context omitted.

> but rather to learn how to conduct research Further, I always assumed PhD level of knowledge meant coming up with the right questions. I would say it is at best a "Lazy Knowledge Rich worker", it won't explore hypothesis if you don't * ask it* to. A PHD would ask those questions to *themselves*. Let me give you a simple example: The other day Claude Code(Max Pro Subscription) commented out a bunch of test assertion…

It's ultrathink one word, not ultra-think . (See below). I use Claude Code with Opus, and had same experience - was pushing it hard to implement complex test, and it gave me an empty test function with test plan inside in a comment (lol). I do want to try Gemini 2.5 Pro, but I don't know a tool which would make experience compatible to Claude Code. Would it make sense to use with Cursor? Do they try to limit context?…

Google has gemini-cli that is pretty close to Claude Code in terms of experience https://github.com/google-gemini/gemini-cli and has a generous free tier. Claude Code is still superior in my experience, Gemini CLI can go off-course pretty quickly if you accept auto edits. But it is handy for code reviews and planning with it's large context window.

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

#318

Post by vl »

ramraj07 wrote: post #310
vl wrote: post #309

I use Claude Code with Opus, and article recommends Gemini 2.5 Pro. I want to try it as well, but I don't know a tool which would make experience compatible to Claude Code. Would it make sense to use with Cursor? Do they try to limit context?

My experience, which seems fairly isolated, is that using Gemini's web chat interface and pasting entire sections of my codebase beats any other agent I've seen. Some come close, and some are good with very large files etc, but if you have a decently organized codebase then using gemini like this beats anything else.

I work on established project with medium-size codebase, copying to chat and back is just not practical.

This is why Claude Code rocks - it quite often finds relevant parts itself.

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

#319

Post by rapind »

worldsayshi wrote: post #281
rapind wrote: post #215

Earlier quoted context omitted.

Not an issue and I'll tell you why. If the gains plateau, then there's really no need to make productivity sacrifices here for the societal good, because there's so much competition, and various levels of open models that aren't far behind, that there will be no reason to stick with a hostile and expensive service unless it's tooling stays leaps ahead of the competition. If the gains don't plateau, well then we're ob…

> 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.

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

#320

Post by milofeynman »

skippyboxedhero wrote: post #52

Have used Claude's GitHub action quite a bit now (10-20 issue implementations, a bit more PR reviews), and it is hit and miss so agree with the enhanced coding rather than just letting it run loose. When the change is very small, self-contained feature/refactor it can mostly work alone, if you have tests that cover the feature then it is relatively safe (and you can do other stuff because it is running in an action,…

The problem here is the infrastructure required to demo the changes to the user. Like yeah you made a code-change, but now I have to pull it, maybe setup data to get it in the right state, check if it's functioning how I want it to. Looking at the code it produced in a diff can waste a lot of your time if it doesn't even work as expected.
Post reply on HN