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

131–140 of 239 posts

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

#131

Post by sebastiennight »

dheera wrote: post #105
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 .

I think we've generally run out of names to give projects and need to start reusing names. Maybe use letters to disambiguate them. Flux A is the ML library Flux B is the T2I model Flux C is the React library Flux D is the physics concept of power per unit area Flux E is the goo you put on solder

Don't forget Fl.ux, which was a very popular way to make "night shift" happen for more than a decade

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

#132

Post by djbusby »

mlboss wrote: post #113

These venture funded startups keep releasing models for free without a business model in sight. I am all for open source but worry it is not sustainable long term.

It's not. These VC firms currently just blasting money+AI at lots of things. Exploring to find what sticks. Expensive discovery.

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

#133

Post by ZoomerCretin »

minimaxir wrote: post #96
ZoomerCretin wrote: post #90

Anyone know why text-to-image models have so many fewer parameters than text models? Are there any large image models (>70b, 400b, etc)?

Diffusion is very efficient encoding/decoding. The only reason that diffusion isn't used for text is because text requires discrete outputs.

Thank you for the explanation!

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

#134

Post by CGamesPlay »

TheAceOfHearts wrote: post #125
mft_ wrote: post #110

Earlier quoted context omitted.

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.

I mean, you can use a fork to make whipped cream, but it won't be easy and it's not the right tool for the job. Does that mean that the fork is useless?

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

#137

Post by refulgentis »

viraptor wrote: post #121
refulgentis wrote: post #120

Earlier quoted context omitted.

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)

Right, that (the black bars) gives no info on how the model works. Thus, you'd love to "know more". ;)

Rest is groping for a reason to make "model is censored [classifier made POST return black image instead of boobs]" something sensical.

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

#138

Post by TheAceOfHearts »

CGamesPlay wrote: post #134
TheAceOfHearts wrote: post #125

Earlier quoted context omitted.

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.

I mean, you can use a fork to make whipped cream, but it won't be easy and it's not the right tool for the job. Does that mean that the fork is useless?

I never said it was useless, just that it fails at this specific problem. One of my complaints with many of these image generation tools is that there's not much communication as to what should be expected from them, nor do they explain the areas where they're expected to succeed or fail.

Recently Claude began to allow generation of SVG drawings, and asking it to draw a unicorn and later add extra tails or horns worked correctly.

A fork exists in physical space and it's pretty intuitive to understand what it can do. These models exist within digital space and are incredibly opaque by comparison.

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

#139

Post by bbkane »

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.

Thanks for the link, and this model looks really good! I think I can tweak the output to make a nice logo [0] for my project!

[0] https://github.com/bbkane/envelope/issues/44

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

#140

Post by thequadehunter »

teamspirit wrote: post #95
Vinnl wrote: post #89

Earlier quoted context omitted.

> (available without sign-in) FLUX.1 [schnell] (Apache 2.0, open weights, step distilled): https://fal.ai/models/fal-ai/flux/schnell Well, I was wondering about bias in the model, so I entered "a president" as the prompt. Looks like it has a bias alright, but it's even more specific than I expected...

You weren’t kidding. Tried three times and all three were variations of the same[0]. [0] https://fal.media/files/elephant/gu3ZQ46_53BUV6lptexEh.png

The interesting part is that it does a good Joe Biden, but Trump always looks weird and alien.

https://imgur.com/a/fgf6Jt3

Post reply on HN