Earlier quoted context omitted.
Then let he who hath not sinned cast the first stone.
Yes scale is totally irrelevant, that's what all of FAANG tell us too :/
Guy who gets his morality from big company internal policies.
Releasing weights for FLUX.1 Krea
101–110 of 112 posts
Re: Releasing weights for FLUX.1 Krea
#102Earlier quoted context omitted.
Likely the same way visual artists ensure that they only learn from images with permissive licenses.
Human learning and computer processing millions of works are different things. I don't think any human artists have seen as many images as the developers used for training.
Only if you don't count artists with able eye(s) living their life?
Re: Releasing weights for FLUX.1 Krea
#103I've never gotten one to make what I am thinking of: A Galton board. At the top, several inches apart are two holes from which balls drop. One drops blue balls, the other red balls. They form a merged distribution below in columns, demonstrating dual overlapping normal distributions Imagine one of these: https://imgur.com/a/DiAOTzJ but with two spouts at the top dropping different colored balls Its attempts: https://…
Have you tried building one irl? I can't find a video of a double one
I have not. It definitely is not something in training sets :)
Re: Releasing weights for FLUX.1 Krea
#104Earlier quoted context omitted.
Human learning and computer processing millions of works are different things. I don't think any human artists have seen as many images as the developers used for training.
Only if you don't count artists with able eye(s) living their life?
Looking at real things while living life is categorically different from viewing millions of artworks made by other artists. I can guarantee you that no artist has ever seen millions of artworks. This comment is not even bad-faith, it's just wrong.
Re: Releasing weights for FLUX.1 Krea
#105Hi! I'm lead researcher on Krea-1. FLUX.1 Krea is a 12B rectified flow model distilled from Krea-1, designed to be compatible with FLUX architecture. Happy to answer any technical questions :)
I recently ran a training experiment using the same dataset, number of steps, and epochs on both Flux Dev and Flux Krea models.
What stood out to me was that Flux Dev followed the text prompts more accurately, whereas Krea’s generations were more loosely aligned or "off" in terms of prompt fidelity with deformations in body type and the architecture.
Does this suggest that Flux Krea requires more training to achieve strong text-to-image alignment compared to Flux Dev? Or is it possible that Krea is optimized differently (e.g. for style, detail, or artistic variation rather than strict prompt adherence)?
Curious if anyone else has experienced this or has any insight into the differences between these two. Would love to hear your thoughts
Re: Releasing weights for FLUX.1 Krea
#106I recently ran a training experiment using the same dataset, number of steps, and epochs on both Flux Dev and Flux Krea models.
What stood out to me was that Flux Dev followed the text prompts more accurately, whereas Krea’s generations were more loosely aligned or "off" in terms of prompt fidelity with deformations in body type and the architecture.
Does this suggest that Flux Krea requires more training to achieve strong text-to-image alignment compared to Flux Dev? Or is it possible that Krea is optimized differently (e.g. for style, detail, or artistic variation rather than strict prompt adherence)?
Curious if anyone else has experienced this or has any insight into the differences between these two. Would love to hear your thoughts
Re: Releasing weights for FLUX.1 Krea
#107Earlier quoted context omitted.
Only if you don't count artists with able eye(s) living their life?
Looking at real things while living life is categorically different from viewing millions of artworks made by other artists. I can guarantee you that no artist has ever seen millions of artworks. This comment is not even bad-faith, it's just wrong.
Are you telling me artists never get inspired by something they saw living their life? Only allowed to be inspired when specifically looking at other art?
Re: Releasing weights for FLUX.1 Krea
#108Earlier quoted context omitted.
Looking at real things while living life is categorically different from viewing millions of artworks made by other artists. I can guarantee you that no artist has ever seen millions of artworks. This comment is not even bad-faith, it's just wrong.
Are you telling me artists never get inspired by something they saw living their life? Only allowed to be inspired when specifically looking at other art?
You must not have responded to the right comment, because your comment is completely irrelevant to mine, as anyone with eyes and basic reading comprehension can see that I never said anything like that.
Re: Releasing weights for FLUX.1 Krea
#109Earlier quoted context omitted.
Are you telling me artists never get inspired by something they saw living their life? Only allowed to be inspired when specifically looking at other art?
You must not have responded to the right comment, because your comment is completely irrelevant to mine, as anyone with eyes and basic reading comprehension can see that I never said anything like that.
> Looking at real things while living life is categorically different from viewing millions of artworks made by other artists
I don't see a difference.
Re: Releasing weights for FLUX.1 Krea
#110Earlier quoted context omitted.
You must not have responded to the right comment, because your comment is completely irrelevant to mine, as anyone with eyes and basic reading comprehension can see that I never said anything like that.
> Looking at real things while living life is categorically different from viewing millions of artworks made by other artists I don't see a difference.
You're factually incorrect, then. Objectively, those things are different, along every single relevant axis - from the psychological, artistic, moral, economic, and legal perspectives.
The vast majority of humans (>99.99999%) can tell that those things are different. The fact that you can't should be somewhat concerning.