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

Gemini 2.5 Flash Image

developers.googleblog.com

Post reply on HN
Print view View original

251–260 of 504 posts

Re: Gemini 2.5 Flash Image

#251

Post by vunderba »

elorant wrote: post #26

I have a certain use case for such image generators. Feed them an entire news article I fetch from bbc and ask it to create an image to accompany the article. Thus far only midjourney managed to understand context. And now this, which is even more impressive. We live in interesting times.

I think most of the SOTA models could probably handle this but you'd probably get better results using a pipeline:

1. Reduce article to a synopsis using an LLM

2. Generate 4-5 varying description prompts from the synopsis

3. Feed the prompts to an imagegen model

Though I'd wager that gpt-image-1 (in the ChatGPT) being multimodal could probably managed it as well.

Re: Gemini 2.5 Flash Image

#252

Post by arresin »

dvh wrote: post #236
pil0u wrote: post #217

Earlier quoted context omitted.

I got scammed similarly (although $10, because I tested first), because 1. it was on YouTube, on a channel called "SpaceX" with verified logo 2. with hundreds of thousands of viewers live 3. with a believable speech from Mr. Musk standing next to its rockets (and knowing his interest in cryptocurrencies). This happened as I was genuinely searching for the actual live stream of SpaceX. I am ashamed, even more so becau…

I am flabbergasted that you both get scammed. I would understand if this was two years ago, but now? Do people really not know about these scams? I can already see down votes coming for victim blaming, but this is to me really shocking. Notice that there isn't "tell hn: don't get scammed by deep fake crypto Elon" because people who usually posts also consider this general knowledge. That's why it's so effective I gue…

Hey it takes courage to admit to it. That’s admirable.

Re: Gemini 2.5 Flash Image

#253

Post by mh- »

asadm wrote: post #226

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.

Can you expand on that? What would ideal look like to you?

Re: Gemini 2.5 Flash Image

#257

Post by esamust »

Strange. I was excited to play around with the 2.5 flash image after testing the nano banana in LMarena, but the results are not at all the same? So I went back to LMarena to replicate my earlier tests but it's way worse than when it was nano banana? Did I miss something?

Re: Gemini 2.5 Flash Image

#258

Post by stefs »

SoKamil wrote: post #220
__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.

If you commented it a decade ago, I would say that at least you own the program and skills in case Google decides to turn off the lights or ask prohibitive price tag. Now you need to pay subscription for PS and maybe there would be some decent open weight model released.

qwen3 is open weights and offers passable image generation

Re: Gemini 2.5 Flash Image

#259

Post by krackers »

echelon wrote: post #114
fariszr wrote: post #60

This is the gpt 4 moment for image editing models. Nano banana aka gemini 2.5 flash is insanely good. It made a 171 elo point jump in lmarena! Just search nano banana on Twitter to see the crazy results. An example. https://x.com/D_studioproject/status/1958019251178267111

> This is the gpt 4 moment for image editing models. No it's not. We've had rich editing capabilities since gpt-image-1, this is just faster and looks better than the (endearingly? called) "piss filter". Flux Kontext, SeedEdit, and Qwen Edit are all also image editing models that are robustly capable. Qwen Edit especially. Flux Kontext and Qwen are also possible to fine tune and run locally. Qwen (and its video gen s…

I'm confused as well, I thought gpt-image could already do most of these things, but I guess the key difference is that gpt-image is not good for single point edits. In terms of "wow" factor it doesn't feel as big as gpt 3->4 though, since it sure _felt_ like models could already do this.
Post reply on HN