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

Gemini 3

blog.google

Post reply on HN
Print view View original

341–350 of 1001 posts

Re: Gemini 3

#341

Post by dktp »

8note wrote: post #319
dktp wrote: post #281

Earlier quoted context omitted.

Google actually changed it somewhat recently (3 months ago, give or take) and you can use Gemini CLI with the "regular" Google AI Pro subscription (~22eur/month). Before that, it required a separate subscription I can't find the announcement anymore, but you can see it under benefits here https://support.google.com/googleone/answer/14534406?hl=en The initial separate subscriptions were confusing at best. Current situ…

how do i actually make it use that though? i got a free year of subscription from buying a phone, but all i get is the free tier in the gemini cli

I also got 1 year through buying my pixel. If you login with the same account through Gemini CLI, it should work (works for me)

However, Gemini CLI is a rather bad product. There is (was?) an issue that makes the CLI fall back to flash very soon in every session. This comment explains it well: https://news.ycombinator.com/item?id=45681063

I haven't used it in a while, except for really minor things, so I can't tell if this is resolved or not

Re: Gemini 3

#342

Post by colechristensen »

slackerIII wrote: post #210

What's the easiest way to set up automatic code review for PRs for my team on GitHub using this model?

Ask it.

If it's good enough to be useful on your code base, it better be good enough to instruct you on how to use it.

How easy it is depends on whether or not they've built that kind of thing in

Re: Gemini 3

#343

Post by qsort »

lairv wrote: post #293

Out of curiosity, I gave it the latest project euler problem published on 11/16/2025, very likely out of the training data Gemini thought for 5m10s before giving me a python snippet that produced the correct answer. The leaderboard says that the 3 fastest human to solve this problem took 14min, 20min and 1h14min respectively Even thought I expect this sort of problem to very much be in the distribution of what the mo…

To be fair a lot of the impressive Elo scores models get are simply due to the fact that they're faster: many serious competitive coders could get the same or better results given enough time.

But seeing these results I'd be surprised if by the end of the decade we don't have something that is to these puzzles what Stockfish is to chess. Effectively ground truth and often coming up with solutions that would be absolutely ridiculous for a human to find within a reasonable time limit.

Re: Gemini 3

#344

Post by xnx »

dwringer wrote: post #236

Well, I tried a variation of a prompt I was messing with in Flash 2.5 the other day in a thread about AI-coded analog clock faces. Gemini Pro 3 Preview gave me a result far beyond what I saw with Flash 2.5, and got it right in a single shot.[0] I can't say I'm not impressed, even though it's a pretty constrained example. > Please generate an analog clock widget, synchronized to actual system time, with hands that upd…

This is cool. Gemini 2.5 Pro was also capable of this. Gemini was able to recreate famous piece of clock artwork in July: https://gemini.google.com/app/93087f373bd07ca2

"Against the Run": https://www.youtube.com/watch?v=7xfvPqTDOXo

Re: Gemini 3

#345

Post by energy123 »

spookie wrote: post #133
aliljet wrote: post #82

Understanding precisely why Gemini 3 isn't front of the pack on SWE Bench is really what I was hoping to understand here. Especially for a blog post targeted at software developers...

Does anyone trust benchmarks at this point? Genuine question. Isn't the scientific consensus that they are broken and poor evaluation tools?

They overly emphasize tasks with small context without noise and red herrings in the context.

Re: Gemini 3

#346

Post by andrewinardeer »

inkysigma wrote: post #145
rvz wrote: post #109

I expect almost no-one to read the Gemini 3 model card. But here is a damning excerpt from the early leaked model card from [0]: > The training dataset also includes: publicly available datasets that are readily downloadable; data obtained by crawlers; licensed data obtained via commercial licensing agreements; user data (i.e., data collected from users of Google products and services to train AI models, along with u…

Isn't Gmail covered under the Workspace privacy policy which forbids using that for training data. So I'm guessing that's excluded by the "in accordance" clause.

The real question is, "For how long?"

Re: Gemini 3

#347

Post by strangescript »

srameshc wrote: post #112

I think I am in this AI fatigue phase. I am past all hype with models, tools and agents and back to problem and solution approach, sometimes code gen with AI , sometimes think and ask for a piece of code. But not offloading to AI and buying all the bs, waiting it to do magic with my codebase.

And by this time next year, this comment is going to look very silly

Re: Gemini 3

#348

Post by thomasahle »

lairv wrote: post #293

Out of curiosity, I gave it the latest project euler problem published on 11/16/2025, very likely out of the training data Gemini thought for 5m10s before giving me a python snippet that produced the correct answer. The leaderboard says that the 3 fastest human to solve this problem took 14min, 20min and 1h14min respectively Even thought I expect this sort of problem to very much be in the distribution of what the mo…

I tried it with gpt-5.1 thinking, and it just searched and found a solution online :p

Re: Gemini 3

#349

Post by storus »

bnchrch wrote: post #87

I've been so happy to see Google wake up. Many can point to a long history of killed products and soured opinions but you can't deny theyve been the great balancing force (often for good) in the industry. - Gmail vs Outlook - Drive vs Word - Android vs iOS - Worklife balance and high pay vs the low salary grind of before. Theyve done heaps for the industry. Im glad to see signs of life. Particularly in their P/E whic…

If you consider surveillance capitalism and dark pattern nudges a good thing, then sure. Gemini has the potential to obliterate their current business model completely so I wouldn't consider that "waking up".

Re: Gemini 3

#350

Post by eevmanu »

xnx wrote: post #184
jhack wrote: post #41

Earlier quoted context omitted.

With this kind of pricing I wonder if it'll be available in Gemini CLI for free or if it'll stay at 2.5.

There's a waitlist for using Gemini 3 for Gemini CLI free users: https://docs.google.com/forms/d/e/1FAIpQLScQBMmnXxIYDnZhPtTP...

In case anyone wants to confirm if this link is official, it is.

https://goo.gle/enable-preview-features

-> https://github.com/google-gemini/gemini-cli/blob/release/v0....

--> https://goo.gle/geminicli-waitlist-signup

---> https://docs.google.com/forms/d/e/1FAIpQLScQBMmnXxIYDnZhPtTP...

Post reply on HN