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

281–290 of 582 posts

Re: Gemini 2.5 Flash

#281

Post by davidsainez »

tkgally wrote: post #254
teleforce wrote: post #245

Earlier quoted context omitted.

>obsequious Thanks for the new word, I have to look it up. "obedient or attentive to an excessive or servile degree" Apparently it means an AI that mindlessly follow your logic and instructions without reasoning and articulation is not good enough.

Another useful word in this context is “sycophancy,” meaning excessive flattery or insincere agreement. Amanda Askell of Anthropic has used it to describe a trait they try to suppress in Claude: https://youtube.com/watch?v=ugvHCXCOmm4&t=10286

The second example she uses is really important. You (used to) see this a lot in stackoverflow where an inexperienced programmer asks how to do some convoluted thing. Sure, you can explain how to do the thing while maintaining their artificial constraints. But much more useful is to say "you probably want to approach the problem like this instead". It is surely a difficult problem and context dependent.

Re: Gemini 2.5 Flash

#282

Post by WiSaGaN »

blueyes wrote: post #229
jeeeb wrote: post #192

Earlier quoted context omitted.

After comparing Gemini Pro and Claude Sonnet 3.7 coding answers side by side a few times, I decided to cancel my Anthropic subscription and just stick to Gemini.

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.

Also the "project" feature in claude improves experience significantly for coder, where you can customize your workflow. Would be great if gemini has this feature.

Re: Gemini 2.5 Flash

#283

Post by jorvi »

arnaudsm wrote: post #133
JeremyNT wrote: post #107

Earlier quoted context omitted.

I use Gemini almost exclusively as a normal user. What am I missing out on that they are far behind on? It seems shockingly good and I've watched it get much better up to 2.5 Pro.

Mostly brand recognition and the earlier Geminis had more refusals. As a consumer, I also really miss the Advanced voice mode of ChatGPT, which is the most transformative tech in my daily life. It's the only frontier model with true audio-to-audio.

> and the earlier Geminis had more refusals.

Its more so that almost every company is running a classifier on their web chat's output.

It isn't actually the model refusing, but rather if the classifier hits a threshold, it'll swap the model's out with "Sorry, let's talk about something else."

This is most apparent with DeepSeek. If you use their web chat with V3 and then jailbreak it, you'll get uncensored output but it is then swapped with "Let's talk about something else" halfway through the output. And if you ask the model, it has no idea its previous output got swapped and you can even ask it build on its previous answer. But if you use the API, you can push it pretty far with a simple jailbreak.

These classifiers are virtually always ran on a separate track, meaning you cannot jailbreak them.

If you use an API, you only have to deal with the inherent training data bias, neutering by tuning and neutering by pre-prompt. The last two are, depending on the model, fairly trivial to overcome.

I still think the first big AI company that has the guts to say "our LLM is like a pen and brush, what you write or draw with it is on you" and publishes a completely unneutered model will be the one to take a huge slice of marketshare. If I had to bet on anyone doing that, it would be xAI with Grok. And by not neutering it, the model will perform better in SFW tasks too.

Re: Gemini 2.5 Flash

#284

Post by energy123 »

blueyes wrote: post #229
jeeeb wrote: post #192

Earlier quoted context omitted.

After comparing Gemini Pro and Claude Sonnet 3.7 coding answers side by side a few times, I decided to cancel my Anthropic subscription and just stick to Gemini.

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.

Google need to fix their Gemini web app at a basic level. It's slow, gets stuck on Show Thinking, rejects 200k token prompts that are sent one shot. Aistudio is in much better shape.

Re: Gemini 2.5 Flash

#285

Post by WiSaGaN »

blahgeek wrote: post #279
simonw wrote: post #267

An often overlooked feature of the Gemini models is that they can write and execute Python code directly via their API. My llm-gemini plugin supports that: https://github.com/simonw/llm-gemini uv tool install llm llm install llm-gemini llm keys set gemini # paste key here llm -m gemini-2.5-flash-preview-04-17 \ -o code_excution 1 \ 'render a mandelbrot fractal in ascii art' I ran that just now and got this: https://g…

> An often overlooked feature of the Gemini models is that they can write and execute Python code directly via their API. Could you elaborate? I thought function calling is a common feature among models from different providers

This common feature requires the user of the API to implement the tool, in this case, the user is responsible to run the code the API outputs. The post you replied suggests that Gemini will run the code for the user behind the API call.

Re: Gemini 2.5 Flash

#286

Post by energy123 »

qwertox wrote: post #190
cjohnson318 wrote: post #164

Earlier quoted context omitted.

Yeah, my wife pays for ChatGPT, but Gemini is fine enough for me.

Just be aware that if you don't add a key (and set up billing) youre granting Google the right to train on your data. To have persons read them and decide how to use them for training.

I thought if you turn off App Activity then that's good enough to protect your data?

Re: Gemini 2.5 Flash

#287

Post by djrj477dhsnv »

Why are most comments here only comparing to Claude and just a few to ChatGPT and none to Grok?

Grok 3 has been my main LLM since its release. Is it not as good as I thought it was?

Re: Gemini 2.5 Flash

#288

Post by jofzar »

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…

My work doesn't have access to 2.5 pro and all these posts are just making me want it so much more.

I hate how slow things are sometimes.

Re: Gemini 2.5 Flash

#289

Post by jofzar »

djrj477dhsnv wrote: post #287

Why are most comments here only comparing to Claude and just a few to ChatGPT and none to Grok? Grok 3 has been my main LLM since its release. Is it not as good as I thought it was?

IMO I will not use Grok while it's owned and related to Elon, not only do I not trust their privacy and data usage (not that I "really" trust open AI/Google etc) I just despise him.

It would have to be very significantly better for me to use it.

Re: Gemini 2.5 Flash

#290

Post by zoogeny »

sumedh wrote: post #270
zoogeny wrote: post #155

Earlier quoted context omitted.

It is more gut feel than a rational or carefully reasoned argument. I think Pichai has been an exceptional revenue maximizer but he lacks vision. I think he is probably capable of squeezing tremendous revenue out of AI once it has been achieved. I like Hassabis in a "good vibe" way when I hear him speak. He reminds me of engineers that I have worked with personally and have gained my respect. He feels less like a pro…

> Altman made a decision to pivot the company towards product which led to the exodus of early research talent. Who was going to fund the research though?

Fair point, and a good reminder not to pass judgement on the actions of others. It is totally possible that Altman made his own prediction of the future and theorized that the only hope he had of competing with the existing big tech companies to realistically achieve an AI for the masses was to show investors a path to profitability.

I should also give Altman a bit more due in that I find his description of a world augmented by powerful AI to be more inspiring than any similar vision I've heard from Pichai.

But I'm not trying to guess their intentions, I am just stating the situation as I see it. And that situation is one where whatever forces have caused it, OpenAI is clearly investing very heavily in product (e.g. windsurf acquisition, even suggesting building a social network). And that shift in focus seems highly correlated with a loss of significant research talent (as well as a healthy dose of boardroom drama).

Post reply on HN