Krea 3 – An image model built for editing at scale

Hello HN,

This is Diego from Krea. Today we are announcing the open beta of Krea 3, our newest foundational image model.

In a nutshell, it’s a frontier image model in terms of quality, but at a fraction of the speed and a fraction of the price.

We will be releasing a more technical update, like we did with Krea 2 (https://www.krea.ai/blog/krea-2-technical-report), but this website shows the capabilities we focused on:

* Native regional edits

* Precise edits so that there is minimal area degradation pixel-wise

* Frontier image quality. Benchmarks still being worked on, but in our internal evals it seems a no-brainer replacement to NBP or GPT Image, both speed and cost-wise.

* Faster edits (AFAICK, this should unlock a pareto frontier for speed in frontier edit models).

In terms of future updates, the team is thinking about letting folks fine-tune Krea 3 (via LoRAs) and open-weights, but we’re still in the middle of launch hiccups and then we’ll revisit the next iterations.

Thanks!

Comparison samples with GPT-Image-2.5 for UI design work:

Starship landing page

Krea 3: https://image.non.io/80b85aaa-286b-4ad7-b3ae-cfab4b491216.we...

GPT-Image-2.5: https://image.non.io/19c5f032-cbc7-4034-9682-3aba451bd669.we...

Fashion history website

Krea 3: https://image.non.io/a6e61219-963c-4921-b453-738960e03308.we...

GPT-Image-2.5: https://image.non.io/c067fd76-9c7e-46c2-a8fd-0b0931bf93f6.we...

Banking website

Krea 3: https://image.non.io/3a1e7b50-36e7-418c-ab47-368ce4f01948.we...

GPT-Image-2.5: https://image.non.io/b96a7e00-3289-4044-90de-e0f86a2ca55e.we...

Overall Krea 3 isn't near GPT-Image-2.5 in terms of visual creativity in this space, but it still should be commended for its ability to render small text details. That said despite it specifying you can add exact hex codes as color reference, I've found that they tend to leak into the design (check the valuation panel of the banking website - notice to the right of the "fixed indexed 7-yr" there's a hex code that made it into the image).

I'd love to get my hands on this to post-train and see if it can be pushed further in this space. I actually see a lot of potential for the model, despite the clear differences in the examples above.

Who told it to add that ugly shadow for the red car?

Can you share any vague details on how much compute/train time was needed to create the model?

Gave it a photo of a woman wearing a T-shirt, asked it to seat her at a fancy bar. It invented a suit jacket, which makes sense, but the drawing on the shirt runs right over the jacket. Gemini did much better.

Great step in the right direction. IMO still to expensive for most “product” use cases.

Fast and affordable imaging editing unlocks tons of cool use cases. It’s refreshing to see latency be held just as important as price as-well!

I’ve been bootstrapping a personal styling app. Funnily enough I’m still seeing best results/$ from grok imagine V1. A now deprecated model.

Are open weights being considered? The one thing I'm always looking forward to is all of the anime fine-tunes.

The pearl earring and the billboard worker didn't feel like frontier quality. Otherwise very impressive

Why is there no comparison with FLUX 3 Image? I am not a big user of image models so maybe those are in different categories (or maybe it's not a relevant frontier model?). Just curious.

Sneaky pricing. Offers you plans when you sign up, if you decline them, it offers you a discount.

Just offer the discounted price up front. If your business cannot survive without sneaky pricing, it's not a good business.

I detest this kind of pricing. It seems that for every subscription I need to threaten to leave to get a discount, and now need an AI to automate that process.

For *image* editing.

The signage text on the fuzzy characters scene was scrambled. It's a no-go for me.

Meh ... there are tons of great paid image models ... where the open weights at!?

Good to know, hope it winds up on model garden soon

[dead]