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

FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs

replicate.com

Post reply on HN
Print view View original

61–70 of 159 posts

Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs

#61

Post by zamadatix »

stoobs wrote: post #42
waffletower wrote: post #21

Earlier quoted context omitted.

Doesn't take a lot of effort to get Flux dev/schnell to run on 3090s unquantized, but I agree that 24gb is the consumer GPU memory limit and there are many with less than that. Flux runs great on modern Mac hardware as well, if you have at least 32gb of unified memory.

I'm running Flux dev fine on a 3080 10GB, unquantised, on windows the nvidia drivers have a function to let it spill over into system ram. It runs a little slower, but it's not a deal-breaker unlike nvidia's pricing and power requirements at the moment

What are you using to run it? When I run Flux Dev in Windows using comfy on a 4090 (24 GB) sometimes it all crashes because it runs out of VRAM when I'm doing too much other stuff.

Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs

#62

Post by sdenton4 »

giancarlostoro wrote: post #9
doctorpangloss wrote: post #4

I'm worried about what happens when more people find out about Ideogram. There are a lot of things that don't appear in ELO scores. For one, they will not reflect that you cannot prompt women's faces in Flux. We can only speculate why.

How locked down is it? My problem with a lot of these is I like to make really ridiculous meme type images, but I run into walls for dumb reasons. Like if I want to make something thats "copyrighted" like a mix of certain characters from one franchise or whatever, I cannot sometimes I get told that the model cannot generate copyrighted content, even though courts ruled that AI generated stuff cannot be copyrighted ei…

It's perfectly happy to make an imperial storm trooper riding a dragon, for what it's worth

Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs

#64

Post by whywhywhywhy »

vessenes wrote: post #28

Flux is so frustrating to me. Really good prompt adherence, strong ability to keep track of multiple parts of a scene, it's technically very impressive. However it seems to have had no training on art-art. I can't get it to generate even something that looks like Degas, for instance. And, I can't even fine tune a painterly art style of any sort into Flux dev. I get that there was working, living artist backlash at SD…

>However it seems to have had no training on art-art. I can't get it to generate even something that looks like Degas, for instance

It feels like they just removed names from the datasets to make it worse at recreating famous people and artists.

Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs

#65

Post by whitehexagon »

I'm running Asahi Linux on a 32GB M1 Pro. Any chance of being able to run text-to-image models locally? I've had some success with LLMs, but only the smaller models. No idea where to start with images, everything seems geared towards msft+nvda.

Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs

#66

Post by zamadatix »

drcongo wrote: post #60
waffletower wrote: post #21

Earlier quoted context omitted.

Doesn't take a lot of effort to get Flux dev/schnell to run on 3090s unquantized, but I agree that 24gb is the consumer GPU memory limit and there are many with less than that. Flux runs great on modern Mac hardware as well, if you have at least 32gb of unified memory.

Really? I tried using it in ComfyUI on my Mac Studio, failed, went searching for answers and all I could find said that something something fp8 can't run on a Mac, so I moved on.

If you're looking for a prebuilt "no tinkering" solution https://diffusionbee.com/ is an open source app (Github link at the bottom of the page if you want to see the code) which has a built in button to import Flux models at the bottom of the home screen.

Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs

#67

Post by collinvandyck76 »

whitehexagon wrote: post #65

I'm running Asahi Linux on a 32GB M1 Pro. Any chance of being able to run text-to-image models locally? I've had some success with LLMs, but only the smaller models. No idea where to start with images, everything seems geared towards msft+nvda.

DiffusionBee will let you do this quite easily.

edit: nevermind, it's a macos app

Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs

#68

Post by miohtama »

loufe wrote: post #52
ilaksh wrote: post #48

Pretty smart model. Here's one I made: https://replicate.com/p/6ez0x8xqvsrga0cjadg8m7bah0

That is astoundingly good adherence to the description. I already liked and was impressed by Flux1 but that is perhaps the most impressive image generation I've ever seen.

Is it going be able to go head-to-head against Midjourney?

Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs

#70

Post by LeoPanthera »

whitehexagon wrote: post #65

I'm running Asahi Linux on a 32GB M1 Pro. Any chance of being able to run text-to-image models locally? I've had some success with LLMs, but only the smaller models. No idea where to start with images, everything seems geared towards msft+nvda.

"Draw Things" is a native Mac app for text to image. It's a a lot more advanced than DiffusionBee, it will download the models for you, and it's free. It's also available for iOS. (!)
Post reply on HN