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

Gemini 2.5 Pro Preview

developers.googleblog.com

Post reply on HN
Print view View original

501–510 of 728 posts

Re: Gemini 2.5 Pro Preview

#501

Post by cap4 »

xyzzy123 wrote: post #493
ssalazar wrote: post #454

Earlier quoted context omitted.

I code with multiple LLMs every day and build products that use LLM tech under the hood. I dont think we're anywhere near LLMs being good at code design. Existing models make _tons_ of basic mistakes and require supervision even for relatively simple coding tasks in popular languages, and its worse for languages and frameworks that are less represented in public sources of training data. I am _frequently_ having to t…

I can't point to any evidence. Also I can't think of what direct evidence I could present that would be convincing, short of an actual demonstration? I would like to try to justify my intuition though: Seems like the key question is: should we expect AI programming performance to scale well as more compute and specialised training is thrown at it? I don't see why not, it seems an almost ideal problem domain? * Short…

This is correct. No idea how people don't see this trend or consider it

Re: Gemini 2.5 Pro Preview

#502

Post by bruce511 »

ookblah wrote: post #497
segphault wrote: post #18

My frustration with using these models for programming in the past has largely been around their tendency to hallucinate APIs that simply don't exist. The Gemini 2.5 models, both pro and flash, seem significantly less susceptible to this than any other model I've tried. There are still significant limitations, no amount of prompting will get current models to approach abstraction and architecture the way a person doe…

the future is probably something that looks pretty "inefficient" to us but a non-factor for a machine. i sometimes think a lot of our code structure is just for our own maintenance and conceptualization (DRY, SRP), but if you throw enough compute adn context at a problem im sure none of this even matters (as much). at least for 90% of the CRUD apps out there, you can def abstract away the entire base framework of get…

I've spent my career writing code in a language which already abstracts 90% of a CRUD-type app away. Indeed there are a whole subset of users who literally don't write a line of code. We've had this since the very early 90's for DOS.

Of course that last 10% does a lot of heavy lifting. Domain expertise, program and database design, sales, support, actually processing the data for more than just simple reports, and so on.

And sure, the code is not maximally efficient in all cases, but it is consistent, and deterministic. Which is all I need from my code generator.

I see a lot of panic from programmers (outside our space) who worry about their futures. As if programming is the ultimate career goal. When really, writing code is the least interesting, and least valuable part of developing software.

Maybe LLMs will code software for you. Maybe they already do. And, yes, despite their mistakes it's very impressive. And yes, it will get better.

But they are miles away from replacing developers- unless your skillset is limited to "coding" there's no need to worry.

Re: Gemini 2.5 Pro Preview

#503

Post by muzani »

ranyume wrote: post #11

I don't know if I'm doing something wrong, but every time I ask gemini 2.5 for code it outputs SO MANY comments. An exaggerated amount of comments. Sections comments, step comments, block comments, inline comments, all the gang.

It's trained on the Google style I guess. Google code always feels excessively commented, to the point where I delete comments from Google samples so I can read the code.

Re: Gemini 2.5 Pro Preview

#505

Post by alex1138 »

Tainnor wrote: post #175
segphault wrote: post #18

My frustration with using these models for programming in the past has largely been around their tendency to hallucinate APIs that simply don't exist. The Gemini 2.5 models, both pro and flash, seem significantly less susceptible to this than any other model I've tried. There are still significant limitations, no amount of prompting will get current models to approach abstraction and architecture the way a person doe…

I definitely get more use out of Gemini Pro than other models I've tried, but it's still very prone to bullshitting. I asked it a complicated question about the Scala ZIO framework that involved subtyping, type inference, etc. - something that would definitely be hard to figure out just from reading the docs. The first answer it gave me was very detailed, very convincing and very wrong. Thankfully I noticed it myself…

The fact that SO much is only discovered after the fact by asking it "Are you sure?" is just insane

There has to be some kind of recursive error checking thing, or something

Re: Gemini 2.5 Pro Preview

#506

Post by snthpy »

I find the naming confusing. Haven't I already been using Gemini 2.5 Pro Preview for the past month? Or was that Experimental?

Also how do i understand the OpenAI model names? I don't use OpenAI anymore since Ilya left but when looking at the benchmarks I'm constantly confused by their model names. We have semantic versioning - why do I need an AI or web search to understand your model name?

Re: Gemini 2.5 Pro Preview

#507

Post by user_7832 »

simonw wrote: post #416

Here's a summary of the 394 comments on this post created using the new gemini-2.5-pro-preview-05-06. It looks very good to me - well grouped, nicely formatted. https://gist.github.com/simonw/7ef3d77c8aeeaf1bfe9cc6fd68760... 30,408 input, 8,535 output = 12.336 cents. 8,500 is a very long output! Finally a model that obeys my instructions to "go long" when summarizing Hacker News threads. Here's the script I used: htt…

> Finally a model that obeys my instructions to "go long"

Something I have found is that when you give a reason for it, LLMs are more likely to do it. The first time I was in a metro with limited internet I told Claude (3.5) as such and asked it to reply in more detail as I could not type back and forth frequently… and it delivered extremely well. Since then I’ve found it to be a helpful prompt for all LLMs I’ve used it on.

Re: Gemini 2.5 Pro Preview

#508

Post by EGreg »

jstummbillig wrote: post #314
segphault wrote: post #18

My frustration with using these models for programming in the past has largely been around their tendency to hallucinate APIs that simply don't exist. The Gemini 2.5 models, both pro and flash, seem significantly less susceptible to this than any other model I've tried. There are still significant limitations, no amount of prompting will get current models to approach abstraction and architecture the way a person doe…

> no amount of prompting will get current models to approach abstraction and architecture the way a person does I find this sentiment increasingly worrisome. It's entirely clear that every last human will be beaten on code design in the upcoming years (I am not going to argue if it's 1 or 5 years away, who cares?) I wished people would just stop holding on to what amounts to nothing, and think and talk more about wha…

Bro. Nothing can be done. What are you talking about? Humans will be replaced for everything, humor, relationships, even raising their own kids, everything can be trained and the AIs just keep improving.

Re: Gemini 2.5 Pro Preview

#509

Post by danpalmer »

cma wrote: post #496
danpalmer wrote: post #492

Earlier quoted context omitted.

From the linked tweet the author seems to be using Gemini through another layer called OpenRouter - it seems quite possible that the issue around lack of clarity of billing/caching could be from that extra layer of indirection.

OpenRouter lets you fund a wallet and spend no more than that. Google will let it go out of control and they purposely delay the billing console by up to 24 hours so if you don't track it all yourself you can get hit big, especially if it is a coding error that uses up to the rate limits.

Well OpenRouter is also facading the API calls, so you may not get the full details of the response back from the upstream LLM service. As far as I can tell the Gemini API returns the token counts in its response enabling you to estimate billing yourself if you want to.

> they purposely delay the billing console by up to 24 hours

This is about scalability and performance. Billing for as many requests per second as a cloud provider gets can't be done live, without significant performance and reliability degradation.

Re: Gemini 2.5 Pro Preview

#510

Post by martinsnow »

byearthithatius wrote: post #367
martinsnow wrote: post #279

Earlier quoted context omitted.

Bwoah it's almost as if react and tailwind is the bees knees ind frontend atm

Sadly. Tailwind is so oof in my opinion. Lets import megabytes just so we don't have to write 5 whole CSS classes. I mean just copy paste the code. Don't get me stared on how ugly the HTML becomes when most tags have 20 f*cking classes which could have been two.

You're doing it wrong. Tailwind is endlessly customizable and after compilation is only kilobytes. But yes lets complain because we don't understand the tooling....
Post reply on HN