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

131–140 of 159 posts

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

#131

Post by reissbaker »

throw14082020 wrote: post #130
davidbarker wrote: post #115

Earlier quoted context omitted.

It's very impressive. I aim for around 50 images if I'm training a style, but only 10 to 20 if training a concept (like an object or a face). I have a MacBook Air so I train using the various API providers. For training a style, I use Replicate: https://replicate.com/ostris/flux-dev-lora-trainer/train For training a concept/person, I use fal: https://fal.ai/models/fal-ai/flux-lora-fast-training With fal, you can trai…

$2 for 2 minutes? Can't you get less than $2 for 1 hour using GPU machines from providers like runpod or AirGPU? I found it a bit expensive to use replicate and fal after 10 minutes of prompting. I have not used runpod or airgpu, and not affiliated.

Yes, renting raw compute via Runpod and friends will generally be much cheaper than renting a higher level service that uses that compute e.g. fal.ai or Replicate. For example, an A6000 on fal.ai is a little over $2/hr (they only show you the price in seconds, perhaps to make it more difficult to compare with ordinary GPU providers); on Runpod an A6000 is less than half that, $0.76/hr in their managed "Secure Cloud." If you're willing to take some risk of boxes disappearing, and don't need much security, Runpod's "Community Cloud" is even cheaper at $0.49/hr.

Similar deal with Replicate: an A100 there is over $5/hr, whereas on Runpod it's $1.64/hr.

And if you use the "serverless" services, the pricing becomes even more astronomical; as you note, $1/minute is unreasonably expensive: that's over 20x the cost of renting 8xH100s on Runpod's "Secure Cloud" (and 8xH100s are extreme overkill for finetuning image generators: even 1xH100 would be sufficient, meaning it's actually 160x markup).

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

#132

Post by davidbarker »

vessenes wrote: post #127
davidbarker wrote: post #124

Earlier quoted context omitted.

Out of curiosity, what do you think of these? https://imgur.com/a/8p7RlMe

Significantly better - they feel like watercolor more than degas but if that’s flux I am impressed!

Unfortunately, not Flux. They're from Midjourney, using a few Degas as a style reference.

Whatever they're doing at Midjourney is still impressive. No training needed and a better result.

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

#133

Post by DeathArrow »

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…

There are people who undistilled Flux so it can be further finetuned, so adding art training won't be an issue.

https://huggingface.co/nyanko7/flux-dev-de-distill

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

#135

Post by TobTobXX »

nubinetwork wrote: post #103

I tried using schnell, it won't fit in a 16gb GPU, and I couldn't get it to run on CPU.

I've sucessfully run schnell and dev on a 12G GPU. They do take 40s/60s repectively, but it works. I used ComfyUI and didn't have to tweak anything.

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

#139

Post by weebull »

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…

I wonder if part of the reason it's good is because it's been trained for a more specific task. I can only imagine that if your concept of a "house" includes range from a stately home to "a pineapple under the sea" you're going to end up with a very generalised concept. It's then takes specific prompting to remove the influences you're not interested in.

I suspect the same goes for art styles. There's such huge variety that really they'd be better surveys by separate models.

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

#140

Post by fortran77 »

doctorpangloss wrote: post #19
liuliu wrote: post #12

Earlier quoted context omitted.

What do you mean? FLUX.1 prompts women or women faces just fine? Do you mean the skin texture is unrealistic or some other artifacts?

Flux will not adhere to your detailed description of a woman's face nearly as well as it does for a man, and it doesn't adhere to text descriptions of faces well in general. This is not a technical limitation, this was a choice in the captioning of the model's dataset and maybe other more sophisticated decisions like loss. It exhibits similar flaws with its representation of male versus female celebrities; it also ex…

I found Flux will barely pay attention to a celebrity name. I like Flux but it makes all realistic human men and women look the same. I tried using celebrity names and it barely made a difference.
Post reply on HN