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

FLUX is fast and it's open source

replicate.com

Post reply on HN
Print view View original

111–120 of 127 posts

Re: FLUX is fast and it's open source

#111

Post by dredmorbius »

starfezzy wrote: post #59
trickstra wrote: post #44

Non-commercial is not open-source, because if the original copyright holder stops maintaining it, nobody else can continue (or has to work like a slave for free). Open-source is about what happens if the original author stops working on it. Open-source gives everyone the license to continue developing it, which obviously means also the ability to get paid. Don't call it open-source if this aspect is missing. Only the…

Doesn’t open source mean the source is viewable/inspectable? I don’t know any closed source apps that let you view the source.

"Open Source" has a specific definition, created by the Open Source Initiative:

https://opensource.org/osd>

Certain usages may be covered by trademark protection, as an "OSI Approved License":

https://opensource.org/trademark-guidelines>

It's based on the Debian Free Software Guidelines (DFSG), which were adopted by the Debian Project to determine what software does, and does not, qualify to be incorporated into the core distribution. (There is a non-free section, it is not considered part of the core distribution.)

https://www.debian.org/social_contract#guidelines>

Both definitions owe much to the Free Software Foundation's "Free Software" definition and the four freedoms protected by the GNU GPL:

- the freedom to use the software for any purpose,

- the freedom to change the software to suit your needs,

- the freedom to share the software with your friends and neighbors, and

- the freedom to share the changes you make.

https://www.gnu.org/licenses/quick-guide-gplv3>

https://www.gnu.org/philosophy/free-sw.html>

Re: FLUX is fast and it's open source

#112

Post by pajeets »

jncfhnb wrote: post #98
pajeets wrote: post #57

Earlier quoted context omitted.

wow this is wild! https://i.postimg.cc/vT6SV7pq/replicate-prediction-6ap8z1jv5... https://i.postimg.cc/vZzMTM71/replicate-prediction-7r4b4p6sj... https://i.postimg.cc/rs6wM5LJ/replicate-prediction-d8s4c93v5... I DEMAND TO KNOW HOW RUN LOCAL SAAR

I’m not sure what saar means here but these images are fairly standard and a drop in the bucket compared to the hideous number of porn fine tunes published daily on civit ai if that’s what you’re looking for

wait you think these images are pornographic?

Re: FLUX is fast and it's open source

#113

Post by Scrapemist »

lolinder wrote: post #15
Vt71fcAqt7 wrote: post #11

Earlier quoted context omitted.

>Please don't complain about tangential annoyances—e.g. article or website formats, name collisions, or back-button breakage. They're too common to be interesting.

In general I think that's true and agree that minor name collision commentary is uninteresting, but in this case we're talking about 11 collisions (and counting) in tech alone, 3 of those in AI/ML and 1 of those specifically in image generation. When it's that bad I think that the frequency of collisions for this name is an interesting topic in its own right.

No

Re: FLUX is fast and it's open source

#114

Post by pajeets »

jsemrau wrote: post #101
slig wrote: post #84

Earlier quoted context omitted.

Could you share the prompt? Thanks.

The prompt is actually not that interesting. "A hand-drawing of a scientific middle-aged man in front of a white background. The man is wearing jeans and a t-shirt. He is thinking a bubble stating "What's in a ReAct JSON prompt?" In the style of European comic book artists of the 1970s and 1980s." Finding the right seed and model configuration is the more difficult part.

just tried it out and it struggled with the bubble caption and adopting other drawing styles but oh god yes this is awesome because an image like this would take forever for me to do and if even arranging someone to commission it is expensive

starving artists are going to famish now, not sure how to feel about it

Re: FLUX is fast and it's open source

#115

Post by jncfhnb »

pajeets wrote: post #112
jncfhnb wrote: post #98

Earlier quoted context omitted.

I’m not sure what saar means here but these images are fairly standard and a drop in the bucket compared to the hideous number of porn fine tunes published daily on civit ai if that’s what you’re looking for

wait you think these images are pornographic?

No, just guessing what you and frankly 80%+ of the community wants

Re: FLUX is fast and it's open source

#116

Post by Mklomoto »

jsemrau wrote: post #68
earthnail wrote: post #65

Earlier quoted context omitted.

However, the blogs or newspapers or print outlets that used to hire them hired them because you couldn’t- it was a differentiator. That differentiator is gone, and as such won’t pay for it anymore. They’ll just use the same AI as you. This destroys the existing market of the artist. To be clear, my comment isn’t meant as a judgment, just as market analysis.

I think it does not take into consideration how much thought and expertise goes into design work. Have a look at the recent controversy about the live-action shooter "concord" that failed spectacularly mainly due to bad character design. Here are two videos that explain that well. I don't think I would ever be capable of designing with that degree of purpose given a generative AI tool. [1] https://www.youtube.com/wat…

There is plenty of space between black and white.

I have taste just no skill in drawing. I don't need an artist, i need a graphics designer and now i can replace a graphics designer with GenAI.

Plenty of Artists can draw very well, but what they learn in the industry is to learn to draw for someone else in an aligned art style etc. That has nothing to do with Art.

Very few people earn there living with being artists.

Its the same thing with all the other people. Look at masterpieces of woodworkers etc. They look interesting, nice but they normall just work for someone else doing their craft not their art.

Re: FLUX is fast and it's open source

#117

Post by dang »

lolinder wrote: post #18
dang wrote: post #14

Earlier quoted context omitted.

" Please don't complain about tangential annoyances—e.g. article or website formats, name collisions, or back-button breakage. They're too common to be interesting. " https://news.ycombinator.com/newsguidelines.html

I just posted a reply to another person quoting this guideline: > In general I think that's true and agree that minor name collision commentary is uninteresting, but in this case we're talking about 11 collisions (and counting) in tech alone, 3 of those in AI/ML and 1 of those specifically in image generation. > When it's that bad I think that the frequency of collisions for this name is an interesting topic in its o…

That makes sense and you're not wrong - it's just that there's a clear tradeoff in terms of more vs. less interesting conversation. Having that simple rule is a net win for HN.

Re: FLUX is fast and it's open source

#118

Post by slig »

jsemrau wrote: post #101
slig wrote: post #84

Earlier quoted context omitted.

Could you share the prompt? Thanks.

The prompt is actually not that interesting. "A hand-drawing of a scientific middle-aged man in front of a white background. The man is wearing jeans and a t-shirt. He is thinking a bubble stating "What's in a ReAct JSON prompt?" In the style of European comic book artists of the 1970s and 1980s." Finding the right seed and model configuration is the more difficult part.

Thank you!

Re: FLUX is fast and it's open source

#120

Post by tylerchilds »

sorenjan wrote: post #23

Text to image models feels inefficient to me. I wonder if it would be possible and better to do it in separate steps, like text to scene graph, scene graph to semantically segmented image, segmented image to final image. That way each step could be trained separately and be modular, and the image would be easier to edit instead of completely replace it with the output of a new prompt. That way it should be much easie…

disney’s multiplane camera but for ai

compositing.

do this with ai today where each layer you want has just the artifact on top of a green background.

layer them in the order you want, then chroma key them out like you’re a 70s public broadcasting station producing reading rainbow.

the ai workflow becomes a singular, recursive step until your disney frame is complete. animate each layer over time and you have a film.

Post reply on HN