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

FLUX.2 [Klein]: Towards Interactive Visual Intelligence

bfl.ai

Post reply on HN
Print view View original

21–30 of 59 posts

Re: FLUX.2 [Klein]: Towards Interactive Visual Intelligence

#21

Post by SV_BubbleTime »

CamperBob2 wrote: post #17
refulgentis wrote: post #7

Earlier quoted context omitted.

Ahh I see, and Klein is basically a response to Z-Image Turbo, i.e. another 4-8B sized model that fits comfortably on a consumer GPU. It’ll be interesting to see how the NSFW catering plays out for the Chinese labs. I was joking a couple months ago to someone that Seedream 4’s talents at undressing was an attempt to sow discord and it was interesting it flew under the radar. Post-Grok going full gooner pedo, I wonder…

They are underselling Z-Image Turbo somewhat. It's arguably the best overall model for local image generation for several reasons including prompt adherence, overall output quality and realism, and freedom from censorship, even though it's also one of the smallest at 6B parameters. ZIT is not far short of revolutionary. It is kind of surreal to contemplate how much high-quality imagery can be extracted from a model t…

Everything you said is exactly the truth.

However.. I’m already expecting the blowback when a Z-Image release doesn’t wow people like the Turbo finetune does. SDXL hasn’t been out two years yet, seems like a decade.

We’ll see. I’m hopeful that Z works as expected and sets the new watermark. I just am not sure it does it right out the gate.

Re: FLUX.2 [Klein]: Towards Interactive Visual Intelligence

#22

Post by satvikpendem »

psubocz wrote: post #8

> FLUX.2 [klein] 4B The fastest variant in the Klein family. Built for interactive applications, real-time previews, and latency-critical production use cases. I wonder what kind of use cases could be "latency-critical production use cases"?

Local models. I'm not gonna wait 10 min for one image on my computer like I did back in the Stable Diffusion days. And image editing in particular.

Re: FLUX.2 [Klein]: Towards Interactive Visual Intelligence

#23

Post by refulgentis »

SV_BubbleTime wrote: post #18
wmf wrote: post #13

Earlier quoted context omitted.

A few days ago people were replying to every image on Twitter saying "Grok, put him/her/it in a bikini" and Grok would just do it. It was minimum effort, maximum damage trolling and people loved it.

Ah. So, see, this is exactly why I need to check apparently. Personally, I go between “I don’t care at all” and “well it’s not ideal” on AI generations. It’s already too late, but the barrier of entry is a lot lower than it was. But I’m applying a good faith argument where GP does not seem to have intended one.

Reducing it to some people put people in bikinis for a couple days for the lulz is...not quite what happened.

You may note I am no shirking violet, nor do I lack perspective, as evidenced by my notes on Seedream. And fortuitiously, I only mentioned it before being dismissed as bad faith: I could not have foreseen needing to call out as credentials until now.

I don't think it's kind to accuse others of bad faith, as evidence by me not passing judgement on the person you are replying to's description.

I do admit it made my stomach churn a little bit to see how quickly people will other. Not on you, I'm sure I've done this too. It's stark when you're on the other side of it.

Re: FLUX.2 [Klein]: Towards Interactive Visual Intelligence

#24

Post by CamperBob2 »

pajtai wrote: post #19

It cannot create an image of a pogo stick. I was trying to get it to create an image of a tiger jumping on a pogo stick, which is way beyond its capabilities, but it cannot create an image of a pogo stick in isolation.

Those are both good benchmark prompts. Z-Image Turbo doesn't like them either:

Tiger on pogo stick: https://i.imgur.com/lnGfbjy.jpeg

Dunno what this is, but it's not a pogo stick: https://i.imgur.com/OmMiLzQ.jpeg

Nano Banana Pro FTW: https://i.imgur.com/6B7VBR9.jpeg

Re: FLUX.2 [Klein]: Towards Interactive Visual Intelligence

#25

Post by vunderba »

I haven’t gotten around to adding Klein to my GenAI Showdown site yet, but if it’s anything like Z-Image Turbo, it should perform extremely well.

For reference, Z-Image Turbo scored 4 out of 15 points on GenAI Showdown. I’m aware that doesn’t sound like much, but given that one of the largest models, Flux.2 (32b), only managed to outscore ZiT (a 6b model) by a single point and is significantly heavier-weight, that’s still damn impressive.

Local model comparisons only:

https://genai-showdown.specr.net/?models=fd,hd,kd,qi,f2d,zt

Re: FLUX.2 [Klein]: Towards Interactive Visual Intelligence

#26

Post by Auracle »

codezero wrote: post #2

I am amazed, though not entirely surprised, that these models keep getting smaller while the quality and effectiveness increases. z image turbo is wild, I'm looking forward to trying this one out. An older thread on this has a lot of comments: https://news.ycombinator.com/item?id=46046916

Quality is increasing, but these small models have very little knowledge compared to their big brothers (Qwen Image/Full size Flux 2). As in characters, artists, specific items, etc.

Re: FLUX.2 [Klein]: Towards Interactive Visual Intelligence

#27

Post by Auracle »

CamperBob2 wrote: post #17
refulgentis wrote: post #7

Earlier quoted context omitted.

Ahh I see, and Klein is basically a response to Z-Image Turbo, i.e. another 4-8B sized model that fits comfortably on a consumer GPU. It’ll be interesting to see how the NSFW catering plays out for the Chinese labs. I was joking a couple months ago to someone that Seedream 4’s talents at undressing was an attempt to sow discord and it was interesting it flew under the radar. Post-Grok going full gooner pedo, I wonder…

They are underselling Z-Image Turbo somewhat. It's arguably the best overall model for local image generation for several reasons including prompt adherence, overall output quality and realism, and freedom from censorship, even though it's also one of the smallest at 6B parameters. ZIT is not far short of revolutionary. It is kind of surreal to contemplate how much high-quality imagery can be extracted from a model t…

Hold on now. Z-Image Turbo has gotten a lot of hype but it's worse at all of those things other than perhaps looking like it was shot on a cell phone camera than Qwen Image and Flux 2 (the full sized version). Once you get away from photographic portraits of people it quickly shows just how little it can do.

It is, however, small and quick.

Re: FLUX.2 [Klein]: Towards Interactive Visual Intelligence

#28

Post by Auracle »

ollin wrote: post #6
refulgentis wrote: post #5

Earlier quoted context omitted.

I’m a bit confused, both you and another commenter mention something called Z-Image, presumably another Flux model? Your frame of it is speculative, i.e. it is forthcoming. Theirs is present tense. Could I trouble you to give us plebes some more context? :) ex. Parsed as is, and avoiding the general confusion if you’re unfamiliar, it is unclear how one can observe “the way it is looking”, especially if turbo was rele…

Z-Image is another open-weight image-generation model by Alibaba [1]. Z-Image Turbo was released around the same time as (non-Klein) FLUX.2 and received generally warmer community response [2] since Z-image Turbo was faster, also high-quality, and reportedly better at generating NSFW material. The base (non-Turbo) version of Z-Image is not yet released. [1] https://tongyi-mai.github.io/Z-Image-blog/ [2] https://www.r…

Z-Image is roughly as censored as Flux 2, from my very limited testing. It got popular because Flux 2 is just really big and slow. It is, however, great at editing, has an amazing breadth of built in knowledge, and has great prompt adherence.

Z Image got popular because the people stuck with 12GB video cards could still use it, and hell - probably train on it, at least once the base version comes out. I think most people disparaging Flux 2 never tried it as they wouldn't want to deal with how slowly it would work on their system, if they even realize that they could run it.

Re: FLUX.2 [Klein]: Towards Interactive Visual Intelligence

#29

Post by vunderba »

pajtai wrote: post #19

It cannot create an image of a pogo stick. I was trying to get it to create an image of a tiger jumping on a pogo stick, which is way beyond its capabilities, but it cannot create an image of a pogo stick in isolation.

It's a tough test for local models - (gpt-image and NB had zero problems) - the only one that came reasonably close was Qwen-Image

Z-Image / Flux 2 / Hidream / Omnigen2 / Qwen Samples:

https://imgur.com/a/tB6YUSu

This is where smaller models are just going to be more constrained and will require additional prompting to coax out the physical description of a "pogo stick". I had similar issues when generating Alexander the Great leading a charge on a hippity-hop / space hopper.

Re: FLUX.2 [Klein]: Towards Interactive Visual Intelligence

#30

Post by vunderba »

Auracle wrote: post #26
codezero wrote: post #2

I am amazed, though not entirely surprised, that these models keep getting smaller while the quality and effectiveness increases. z image turbo is wild, I'm looking forward to trying this one out. An older thread on this has a lot of comments: https://news.ycombinator.com/item?id=46046916

Quality is increasing, but these small models have very little knowledge compared to their big brothers (Qwen Image/Full size Flux 2). As in characters, artists, specific items, etc.

Agreed - given what Tongyi-MAI Lab was able to accomplish with a 6b model - I would love to see what they could do with something larger. Somewhere in the range of 15-20b, between these smaller models (ZiT, Klein) and the significantly larger models (Flux.2 dev).
Post reply on HN