Gemini 2.5 Flash Image • Hacker News Board

Re: Gemini 2.5 Flash Image

#221

Post by qingcharles » Tue, Aug 26, 2025, 5:22 PM UTC

druskacik wrote: post #211

qingcharles wrote: post #175

Earlier quoted context omitted.

I've been testing it for several weeks. It can produce results that are truly epic, but it's still a case of rerolling the prompt a dozen times to get an image you can use. It's not God. It's definitely an enormous step though, and totally SOTA.

Is it because the model is not good enough at following the prompt, or because the prompt is unclear? Something similar has been the case with text models. People write vague instructions and are dissatisfied when the model does not correctly guess their intentions. With image models it's even harder for model to guess it right without enough details.

No, my prompts are very, very clear. It just won't follow them sometimes. Also this model seems to prefer shorter prompts, in my experience.

Re: Gemini 2.5 Flash Image

#222

Post by Imustaskforhelp » Tue, Aug 26, 2025, 5:23 PM UTC

atrus wrote: post #134

Imustaskforhelp wrote: post #63

Earlier quoted context omitted.

hm, but isn't it wild thinking that elon is talking to you and asking you for 15k , like bro has the money of his lifetime, why would he ask you? It doesn't make that much sense idk

I remember watching the SpaceX channel on youtube, which isn't a legit source. AI Elon basically says "I want to help make bitcoin more popular, let me show you how easy it it to transfer money around with btc. Send my $X and I'll send you back $2X! It's very inline with a typical elon message (I'll give you 1 million to vote R), it's on a channel called SpaceX. It's pretty believable. Granted I played Runescape and…

Now I have never played runescape but have heard of this legendary game in references.

For some reason, my mind confused runescape with neopets from the odd1sout video which I think is a good watch.

Scams That Should be Illegal : https://www.youtube.com/watch?v=XyoBNHqah30

Re: Gemini 2.5 Flash Image

#223

Post by conradkay » Tue, Aug 26, 2025, 5:23 PM UTC

DonHopkins wrote: post #172

knicholes wrote: post #45

Earlier quoted context omitted.

[flagged]

[flagged]

I don't think he's the gullible one, check their bio ;)

Re: Gemini 2.5 Flash Image

#224

Post by casey2 » Tue, Aug 26, 2025, 5:24 PM UTC

4 out of 7 images show a woman 1 out of 7 show a man I feel like this is trying to advertise power over women to men. Which makes it evil.

Re: Gemini 2.5 Flash Image

#225

Post by ctippett » Tue, Aug 26, 2025, 5:24 PM UTC

__rito__ wrote: post #192

I am glad that I never decided to become a photoshop pro. I always contemplated about it, seemed attractive for a while, but glad that I decided against it. RIP r/photoshopbattles. It was in the endless list of new shiny 'skills' that feels good to have. Now I can use nano-banana instead. Other models will soon follow, I am sure.

Interesting take. I'm a programmer, but learned Photoshop in the early 2000s and had a blast making and editing images for fun. Sure, the generative models today can do a far better job than anything I could come up with, but that doesn't detract from the experience and skills I picked up over the years.

If anything, knowing Photoshop (I use Affinity Designer/Photo these days) is actually incredibly useful to finesse the output produced by AI. No regrets.

Re: Gemini 2.5 Flash Image

#226

Post by asadm » Tue, Aug 26, 2025, 5:26 PM UTC

this is amazing. I just wish models would have more non-textual controls. I don't want to TYPE my instructions. We need a better UI for editing images with AI.

Re: Gemini 2.5 Flash Image

#227

Post by yborg » Tue, Aug 26, 2025, 5:26 PM UTC

simianwords wrote: post #83

L like it but it is very restricted. I can't modify people's faces etc.

Yeah, it doesn't look like the Safety Settings actually do anything, it refuses to edit anything with any words that could be construed to have sexual connotation even with restrictions all turned off.

Re: Gemini 2.5 Flash Image

#228

Post by casey2 » Tue, Aug 26, 2025, 5:27 PM UTC

quantumHazer wrote: post #199

echelon wrote: post #196

Earlier quoted context omitted.

Programming and everything else will eventually fall to automation, too. It's just a matter of time. Engineering probably takes a while (5 years? 10 years?) because errors multiply and technical debt stacks up. In images, that's not so much of a big deal. You can re-roll. The context and consequences are small. In programs, bad code leads to an unmaintainable mess and you're stuck with it. But eventually this will ca…

Both of you are wrong and this is not good discussion level for HN

If being wrong isn't good discussion for HN then they should delete the site

Re: Gemini 2.5 Flash Image

#229

Post by AbraKdabra » Tue, Aug 26, 2025, 5:29 PM UTC

umbra07 wrote: post #147

AbraKdabra wrote: post #96

Earlier quoted context omitted.

I don't mean to be rude, but this sounds like natural selection doing its work.

That's the sort of statement that remains extremely rude even if you try and prefix it with "I don't mean to be rude".

It's not rude if it's the truth.

Also he's a troll so...

Re: Gemini 2.5 Flash Image

#230

Post by carimura » Tue, Aug 26, 2025, 5:37 PM UTC

skybrian wrote: post #177

Like most image generators, it didn’t pass the piano keyboard test. (Black keys are wrong.) https://aistudio.google.com/app/prompts?state=%7B%22ids%22:%...

or my "hands with palms facing down" test.... no matter how hard I try it just can't get open hands, palms down.