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

Flux: Open-source text-to-image model with 12B parameters

blog.fal.ai

Post reply on HN
Print view View original

121–130 of 239 posts

Re: Flux: Open-source text-to-image model with 12B parameters

#121

Post by viraptor »

refulgentis wrote: post #120
viraptor wrote: post #109

Censored a bit, but not completely. I can get occasional boobs out of it, but sometimes it just gives the black output.

This gives you no info on how the model works. what is being applied is fal's post-inference "is this NSFW?" filter model So your censorship investigation (via boobs) is testing a completely different, unrelated, model.

It does provide information. Regardless of whether they use a post-inference filter, we now know that the model itself was trained on and can produce NSFW content. Compare this to SD3 which produces a noise pattern if you request naked bodies.

(Also you can download the model itself to check the local behaviour without extra filters. Unfortunately I don't have time to do it right now, but I'd love to know)

Re: Flux: Open-source text-to-image model with 12B parameters

#123

Post by vunderba »

Vast majority of comparisons aren't really putting these new models through their paces.

The best prompt adherence on the market right now BY FAR is DALL-E 3 but it still falls down on more complicated concepts and obviously is hugely censored - though weirdly significantly less censored if you hit their API directly.

I quickly mocked up a few weird/complex prompts and did some side-by-side comparisons with Flux and DALL-E 3. Flux is impressive and significantly performant particularly since both the dev/shnell models have been confirmed by Black Forest to be runnable via ComfyUI.

https://mordenstar.com/blog/flux-comparisons

Re: Flux: Open-source text-to-image model with 12B parameters

#124

Post by codezero »

smusamashah wrote: post #58

Holy crap this is amazing. I saw an image with a prompt on reddit and didn't believe it was generated imaged. I thought it must be joke that people are sharing non-generated images in the thread. Reddit message: https://www.reddit.com/r/StableDiffusion/comments/1ehh1hx/an... Linked image: https://preview.redd.it/dz3djnish2gd1.png?width=1024&format=... The prompt: > Photo of Criminal in a ski mask making a phone call…

I love the background details. It has HALO as a storefront logo!

This model appears to do well with fingers and hands out of the box.

Re: Flux: Open-source text-to-image model with 12B parameters

#125

Post by TheAceOfHearts »

mft_ wrote: post #110
metadat wrote: post #107

Earlier quoted context omitted.

The playground is a drag. After accepting being forced to sign up, attach my GitHub, and hand over my email address, I entered the desired prompt and waited with anticipation.. Only to see a black screen and how much it's going to cost per megapixel. Bummer. After seeing what was generated in the blog post I was excited to try it! Now feeling disappointed. I was hoping it'd be more like https://play.go.dev . Good luc…

https://replicate.com/black-forest-labs/flux-dev is working very nicely. No sign-up.

My go-to test for these tools so far has been the seven horned, seven eyed lamb mentioned in the Book of Revelation. Every tool I've tried has failed at this task.

Re: Flux: Open-source text-to-image model with 12B parameters

#126

Post by harrisonjackson »

vunderba wrote: post #123

Vast majority of comparisons aren't really putting these new models through their paces. The best prompt adherence on the market right now BY FAR is DALL-E 3 but it still falls down on more complicated concepts and obviously is hugely censored - though weirdly significantly less censored if you hit their API directly. I quickly mocked up a few weird/complex prompts and did some side-by-side comparisons with Flux and…

Your comparisons are all with the flux shnell model

> The fastest image generation model tailored for local development and personal use

Versus flux pro or dev models

Re: Flux: Open-source text-to-image model with 12B parameters

#127

Post by vunderba »

harrisonjackson wrote: post #126
vunderba wrote: post #123

Vast majority of comparisons aren't really putting these new models through their paces. The best prompt adherence on the market right now BY FAR is DALL-E 3 but it still falls down on more complicated concepts and obviously is hugely censored - though weirdly significantly less censored if you hit their API directly. I quickly mocked up a few weird/complex prompts and did some side-by-side comparisons with Flux and…

Your comparisons are all with the flux shnell model > The fastest image generation model tailored for local development and personal use Versus flux pro or dev models

I did put them through pro/dev as well just to be safe. The quality changes and you can play with guidance (cranking it all the way to 10) but it doesn't make a significant difference for these prompts from what I could tell.

Several iterations and these were the best I got out of schnell, dev and pro respectively for the following prompt:

"a fantasy creature with the body of a dragon and a beachball for a head, hybrid, best quality, shadows and lighting, fantasy illustration muted"

https://gondolaprime.pw/pictures/schnell-dev-pro.jpg

Re: Flux: Open-source text-to-image model with 12B parameters

#128

Post by shubik22 »

burkaygur wrote: post #32

hi friends! burkay from fal.ai here. would like to clarify that the model is NOT built by fal. all credit should go to Black Forest Labs ( https://blackforestlabs.ai/ ) which is a new co by the OG stable diffusion team. what we did at fal is take the model and run it on our inference engine optimized to run these kinds of models really really fast. feel free to give it a shot on the playgrounds. https://fal.ai/models…

thanks for hosting the model! i created an account to try it out, you started emailing me with “important notice: low account balance - action required” and now it seems like there’s no way for me to unsubscribe or delete my account. is that the case? thanks!

Re: Flux: Open-source text-to-image model with 12B parameters

#129

Post by liuliu »

metadat wrote: post #112
mft_ wrote: post #110

Earlier quoted context omitted.

https://replicate.com/black-forest-labs/flux-dev is working very nicely. No sign-up.

Thanks this one actually works, pretty amazing. Remarkably better than the "DrawThings" iPhone app (my only reference point).

You should expect Flux model support in the next a few days coming to the app.

Re: Flux: Open-source text-to-image model with 12B parameters

#130

Post by djbusby »

refulgentis wrote: post #67
nextos wrote: post #45

Earlier quoted context omitted.

The name is a bit unfortunate given that Julia's most popular ML library is called Flux. See: https://fluxml.ai . This library is quite well known, 3rd most starred project in Julia: https://juliapackages.com/packages?sort=stars . It has been around since, at least, 2016: https://github.com/FluxML/Flux.jl/graphs/code-frequency .

There was a looong distracting thread a month ago about something similar, niche language, might have been Julia, had a package with the same name as $NEW_THING. I hope this one doesn't stir as much discussion. It has 4000 stars, there isnt a large mass of people who view the world through the lens of "Flux is ML library". No one will end up in a "who is on first?" discussion because of it. If this line of argument i…

Like the Go language that existed before Google Go.
Post reply on HN