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

Gemini 2.5 Flash

developers.googleblog.com

Post reply on HN
Print view View original

351–360 of 582 posts

Re: Gemini 2.5 Flash

#351

Post by crossroadsguy »

zoogeny wrote: post #131

Google making Gemini 2.5 Pro (Experimental) free was a big deal. I haven't tried the more expensive OpenAI models so I can't even compare, only to the free models I have used of theirs in the past. Gemini 2.5 Pro is so much of a step up (IME) that I've become sold on Google's models in general. It not only is smarter than me on most of the subjects I engage with it, it also isn't completely obsequious. The model push…

One difference, and imho that’s a big difference — you can’t use any of the Google’s chatbots/models without being logged in, unlike chatgpt.

Re: Gemini 2.5 Flash

#352

Post by zaphirplane »

fsndz wrote: post #205
zoogeny wrote: post #131

Google making Gemini 2.5 Pro (Experimental) free was a big deal. I haven't tried the more expensive OpenAI models so I can't even compare, only to the free models I have used of theirs in the past. Gemini 2.5 Pro is so much of a step up (IME) that I've become sold on Google's models in general. It not only is smarter than me on most of the subjects I engage with it, it also isn't completely obsequious. The model push…

More and more people are coming to the realisation that Google is actually winning at the model level right now.

What’s with the Google cheer squad in this thread, usually it’s Google lost its way and is evil.

Can’t be employees cause usually there is a disclaimer

Re: Gemini 2.5 Flash

#353

Post by rgoulter »

viraptor wrote: post #160
halfmatthalfcat wrote: post #90

Earlier quoted context omitted.

Or we can just learn to write it ourselves in the same amount of time /shrug

If you're going to need scripts like that every week - sure. If you need it once a year on average... not likely. There's a huge amount of things we could learn but do them so infrequently that we outsource it to other people.

Right.

This is one case where I've found writing code with LLMs to be effective.

With some unfamiliar tool I don't care about too much (e.g. GitHub Actions YAML or some build script), I just want it to work, & then focus on other things.

I can spend time to try and come up with something that works; something that's robust & idiomatic.. but, likely I won't be able to re-use that knowledge before I forget it.

With an LLM, I'll likely get just as good a result; or if not, will have a good starting point to go from.

Re: Gemini 2.5 Flash

#354

Post by visarga »

jdthedisciple wrote: post #212
Nihilartikel wrote: post #202

Earlier quoted context omitted.

100% agree. I had Gemini flash 2 chew through thousands of points of nasty unstructured client data and it did a 'better than human intern' level conversion into clean structured output for about $30 of API usage. I am sold. 2.5 pro experimental is a different league though for coding. I'm leveraging it for massive refactoring now and it is almost magical.

> thousands of points of nasty unstructured client data What I always wonder in these kinds of cases is: What makes you confident the AI actually did a good job since presumably you haven't looked at the thousands of client data yourself? For all you know it made up 50% of the result.

In my professional opinion they can extract data at 85-95% accuracy.

Re: Gemini 2.5 Flash

#355

Post by indexerror »

julianeon wrote: post #347
blueyes wrote: post #229

Earlier quoted context omitted.

One of the main advantages Anthropic currently has over Google is the tooling that comes with Claude Code. It may not generate better code, and it has a lower complexity ceiling, but it can automatically find and search files, and figure out how to fix a syntax error fast.

Related: Only Claude (to my knowledge) has a desktop app which can directly, and usually quite intelligently, modify files and create repos on your desktop. It's the only "agentic" option among the major players. "Claude, make me an app which will accept Stripe payments and sell an ebook about coding in Python; first create the app, then the ebook." It would take a few passes but Claude could do this; obviously you c…

OpenAI just released Codex, which is basically the same as Claude Code.

Re: Gemini 2.5 Flash

#357

Post by thrdbndndn »

julianeon wrote: post #347
blueyes wrote: post #229

Earlier quoted context omitted.

One of the main advantages Anthropic currently has over Google is the tooling that comes with Claude Code. It may not generate better code, and it has a lower complexity ceiling, but it can automatically find and search files, and figure out how to fix a syntax error fast.

Related: Only Claude (to my knowledge) has a desktop app which can directly, and usually quite intelligently, modify files and create repos on your desktop. It's the only "agentic" option among the major players. "Claude, make me an app which will accept Stripe payments and sell an ebook about coding in Python; first create the app, then the ebook." It would take a few passes but Claude could do this; obviously you c…

Copilot agent mode?

Re: Gemini 2.5 Flash

#358

Post by visarga »

wkat4242 wrote: post #231
antirez wrote: post #182

Earlier quoted context omitted.

Maybe because Google is largely responsible, paying for the research, of most of the results we are seeing now. I'm not a Google fan, in the web side, and in their idea of what software engineering is, but they deserve to win the AI race, because right now all the other players provided a lot less than what Google did as public research. Also, with Gemini 2.5 PRO, there was a big hype moment, because the model is of…

Maybe they deserve it but it would be really bad for the world. Because they will enshittify the hell out of it once they're established. That's their MO. I don't want Google to have a stranglehold over yet another type of online service. So I avoid them. And things are going so fast now, whatever Google has today that might be better than the rest, in two months the rest will have it too. Of course Google will have…

The situation with LLMs is much different than search, Google doesn't have such a large lead here. LLMs are social things, they learn from each other, any provider with SOTA model will see its abilities leaked through synthetic training data. That's what GPT-4 did for a year, against the wishes of OpenAI, powering up millions of open model finetunes.

Re: Gemini 2.5 Flash

#359

Post by menshiki »

As a person mostly using AI for everyday tasks and business-related research, it's very impressive how quickly they've progressed. I would consider all models before 2.0 totally unusable. Their web interface, however, is so much worse than that of the ChatGPT macOS app.

Re: Gemini 2.5 Flash

#360

Post by FooBarWidget »

jdthedisciple wrote: post #212
Nihilartikel wrote: post #202

Earlier quoted context omitted.

100% agree. I had Gemini flash 2 chew through thousands of points of nasty unstructured client data and it did a 'better than human intern' level conversion into clean structured output for about $30 of API usage. I am sold. 2.5 pro experimental is a different league though for coding. I'm leveraging it for massive refactoring now and it is almost magical.

> thousands of points of nasty unstructured client data What I always wonder in these kinds of cases is: What makes you confident the AI actually did a good job since presumably you haven't looked at the thousands of client data yourself? For all you know it made up 50% of the result.

You can use AI to verify its own work. Last time I split a C++ header file into header + implementation file. I noticed some code got rewritten in a wrong manner, so I asked it to compare the new implementation file against the original header file, but to do so one method at a time. For each method, say whether the code is exactly the same and has the same behavior, ignoring superficial syntax changes and renames. Took me a few times to get the prompt right, though.
Post reply on HN