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

691–700 of 728 posts

Re: Gemini 2.5 Pro Preview

#691

Post by gtirloni »

mark_l_watson wrote: post #666
namaria wrote: post #612

Earlier quoted context omitted.

Exponential over which metric exactly? Training dataset size, compute required yeah these have grown exponentially. But has any measure capability? Because exponentially growing costs with linear or not measurable improvements is not a great trajectory.

Exponential in how useful LLM APIs and LLM based products like Google AI Lab, ChatGPT, etc. are to me personally. I am the data point I care about. I have a pet programming problem that every few months I try to solve with the current tools of the day. I admit this is anecdotal, just my personal experiences. Metrics like training data set size are less interesting now given the utility of smaller synthetic data sets.…

I think you're using words like "exponential" and "exponentially" as intensifiers and not in the mathematical sense, right? People are engaging in discussions with you expecting numbers to back your claims because of that.

Re: Gemini 2.5 Pro Preview

#692

Post by coredog64 »

GTP wrote: post #664
lallysingh wrote: post #617

Earlier quoted context omitted.

The software tool takes a higher-level input to produce the executable. I'm waiting for LLMs to integrate directly into programming languages. The discussions sound a bit like the early days of when compilers started coming out, and people had been using direct assembler before. And then decades after, when people complained about compiler bugs and poor optimizers.

> I'm waiting for LLMs to integrate directly into programming languages. What do you mean? How would this look like in your view?

Not OP, but probably similar to how tool calling is managed: You write the docstring for the function you want, maybe include some specific constraints, and then that gets compiled down to byte code rather than human authored code.

Re: Gemini 2.5 Pro Preview

#693

Post by mosschief »

ArthurStacks wrote: post #551
auggierose wrote: post #539

Earlier quoted context omitted.

Oh, they are definitely capable, I am using them every day, and build my own MCP servers. But you cannot test a distributed system "fully". The only test I believe in is understanding every single line of code myself, or knowing that somebody else does. At this point, I don't trust the AI for anything, although it makes a very valuable assistant. Very soon our AI built software systems will break down in spectacular…

I have no idea why you think you can't test a distributed system. Hopefully you are not in the business of software development. You certainly wouldnt be working at my company. Secondly, people are not just blindly having AI write code with no idea how it works. The AI is acting as a senior consultant helping the developer to design and build the systems and generating parts of the code as they work together.

I'm very confused by this. I have in no way seen AI that can act as a senior consultant to any professional software engineer. I work with AI all the time and am not doubting that it is very useful, but this seems like dreaming to me. It frequently gets confused and doesn't understand the bigger picture, particularly when large contexts are involved. Solving small problems it is often helpful but I can't imagine how anyone could believe it is in any way a replacement for a senior engineer in its current form.

Re: Gemini 2.5 Pro Preview

#694

Post by sanderjd »

skydhash wrote: post #686
sanderjd wrote: post #361

Earlier quoted context omitted.

Yeah I think we largely agree. But I do know people, mostly experienced product managers, who are excited about "vibecoding" expressly as a prototyping / demo creation tool, which can be useful in conjunction with people who know how to turn the prototypes into real software. I'm sure lots of people aren't seeing it this way, but the point I was trying to make about this being a skill differentiator is that I think u…

If you're really prototyping a product, a simple mockup with a tool like Balsamiq can get you quite far for communication and ideation. But more often, when people want a live prototype, it's because they plan to spin some lies as "sales and marketing".

Well we can agree to disagree about this one :)

What I've seen people use it for to, in my opinion, great effect is to demonstrate capabilities that exist, but for which there are many different possibilities for how to combine and present them to users.

Sure, you can just put together a clickable mock up like people have been doing for years, but putting together functional UIs that call out to existing APIs but cobble them together in different ways, that's actually less smoke and mirrors sales spin.

Re: Gemini 2.5 Pro Preview

#695

Post by cma »

danpalmer wrote: post #509
cma wrote: post #496

Earlier quoted context omitted.

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…

With open router, no matter what happens you won't spend more than you deposited or owe more. It's much safer.

> 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.

I don't buy this, for LLMs specifically. For lots of things a cloud provider gives, things might be aggregated and batched before showing up, but there is no reason your LLM spend should take nearly as long as bank system clearing to show. Especially an estimate, which it already gets disclaimed as.

There are companies that will monitor your cloud spend much faster pretty cheaply, and they are essentially having to reimplement the whole thing from the outside and keep up with Google's pricing changes through a shadow recreation of the billing system.

And open router is able to reflect your spend to you immediately, or couldn't implement their cap. If they can do it why can't Google, at least for the broad number of customers without custom price agreements.

Re: Gemini 2.5 Pro Preview

#697

Post by ArthurStacks »

mosschief wrote: post #693
ArthurStacks wrote: post #551

Earlier quoted context omitted.

I have no idea why you think you can't test a distributed system. Hopefully you are not in the business of software development. You certainly wouldnt be working at my company. Secondly, people are not just blindly having AI write code with no idea how it works. The AI is acting as a senior consultant helping the developer to design and build the systems and generating parts of the code as they work together.

I'm very confused by this. I have in no way seen AI that can act as a senior consultant to any professional software engineer. I work with AI all the time and am not doubting that it is very useful, but this seems like dreaming to me. It frequently gets confused and doesn't understand the bigger picture, particularly when large contexts are involved. Solving small problems it is often helpful but I can't imagine how…

[flagged]

Re: Gemini 2.5 Pro Preview

#698

Post by throwup238 »

taf2 wrote: post #623
throwup238 wrote: post #278

Earlier quoted context omitted.

> I really liked the Gemini 2.5 pro model when it was first released - the upload code folder was very nice (but they removed it). Removed from where? I use the attach code folder feature every day from the Gemini web app (with a script that clones a local repo that deletes .git and anything matching a gitignore pattern).

Maybe I got stuck with a bad experiment that removed it but it has been gone for me for a few weeks so I just stopped using it

It just got removed from the Add menu for me too. Now I have to click "Import Code" and then the "Upload Folder" button in the dialog. Maybe you got this roll out much earlier than I did?

Re: Gemini 2.5 Pro Preview

#699

Post by ArthurStacks »

resize2996 wrote: post #689
ArthurStacks wrote: post #680

Earlier quoted context omitted.

[flagged]

weird ad hominem, but you do you. I'm trying to figure out this logical inconsistency: "AI has made my workers more productive, therefore my workers are worth less." My general theory is that there is more than enough engineering work to go around

[flagged]

Re: Gemini 2.5 Pro Preview

#700

Post by weq »

ArthurStacks wrote: post #599
DonHopkins wrote: post #594

Earlier quoted context omitted.

[flagged]

LLMs cannot replace what we do. Only AGI could do that, at which point you could say the same about anything. 'Racist' in your culture, not in mine.

The reason you feel safe now is because of the marketing tactics of AI companies in pushing their phished goods on the world. LLMs have done anything yet other then reduced the barrier of entry into the software field. Like what google search and stackoverflow did 10yrs ago. The same principles apply, if your only skill is using an LLMs (or google searching) then you will be the first replaced when the markets turn. The ability to reason about options of a company in making money over the short term, vs long term, should be fairly easy to reason about based on the availibility of news. AI companies already know this. The stratergy has been played out. They make more money this way. They get to suck up all the info from your corperation, because they will get that data. Once they build these models, they will replace you too. Sure your saving time and money today, but thats just the cost of building the model for them.
Post reply on HN