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

DeepFloyd IF: open-source text-to-image model

github.com

Post reply on HN
Print view View original

231–237 of 237 posts

Re: DeepFloyd IF: open-source text-to-image model

#231

Post by anticensor »

mkaic wrote: post #158
Jackson__ wrote: post #147

Earlier quoted context omitted.

As far as I can tell from Emad's discord and twitter discussion, the idea appears to be to make this a "research" release, and therefore the worse license. At a later point the model will be renamed "StableIf", and released with a similar license to StableDiffusion.

Yeah Emad was clarifying this on the LAION discord the other day — plan is to have a better-licensed version out Eventually™, guess we'll see how long that takes.

for certain values of soon™ and eventually™

Re: DeepFloyd IF: open-source text-to-image model

#232

Post by krisoft »

alex_sf wrote: post #228
krisoft wrote: post #192

Earlier quoted context omitted.

> You can't use this to make logos for any commercial product Yeah good luck figuring out that a particular logo was generated with this particular model. And if someone does good luck doing anything about it. With this amount of fear one wouldn't dare to cross a road without three layers of bubble wrap, plus written authorisation from a lawyer plus a feasibility study from a traffic engineer.

You've never gone through an acquisition or due diligence, have you?

Clearly.

Explain it to me what will happen with the generated logo at an acquisition or due diligence.

Re: DeepFloyd IF: open-source text-to-image model

#233

Post by alex_sf »

krisoft wrote: post #232
alex_sf wrote: post #228

Earlier quoted context omitted.

You've never gone through an acquisition or due diligence, have you?

Clearly. Explain it to me what will happen with the generated logo at an acquisition or due diligence.

You'll be asked to produce documents verifying your ownership and/or compliance with any licenses for all intellectual property. That includes code and graphics (logos).

Re: DeepFloyd IF: open-source text-to-image model

#234

Post by mkaic »

causality0 wrote: post #224
dragonwriter wrote: post #216

Earlier quoted context omitted.

The VRAM requirements are higher (14GB) so lots of things that can do SD won’t do this with thr existing toolchain. But some of that is “aftermarket” SD optimization, and this maybe could see some of that, too. But there are consumer cards with 14GB+ VRAM, so its not, even before optimization, out of reach of consumer hardware.

Damn. That's basically just the 4090 and 4080.

The 3090 has 24GB so it's an option as well.

Re: DeepFloyd IF: open-source text-to-image model

#235

Post by krisoft »

alex_sf wrote: post #233
krisoft wrote: post #232

Earlier quoted context omitted.

Clearly. Explain it to me what will happen with the generated logo at an acquisition or due diligence.

You'll be asked to produce documents verifying your ownership and/or compliance with any licenses for all intellectual property. That includes code and graphics (logos).

Sure. And what documents and paperwork do you expect if I the owner of the company drawn the logo myself with a bit of a crayon/inkscape/gimp/blender? Those exact documents will be produced.

Re: DeepFloyd IF: open-source text-to-image model

#236

Post by alex_sf »

krisoft wrote: post #235
alex_sf wrote: post #233

Earlier quoted context omitted.

You'll be asked to produce documents verifying your ownership and/or compliance with any licenses for all intellectual property. That includes code and graphics (logos).

Sure. And what documents and paperwork do you expect if I the owner of the company drawn the logo myself with a bit of a crayon/inkscape/gimp/blender? Those exact documents will be produced.

I would expect a lawsuit if you gave them to investors, and jailtime if it was part of an IPO. Keep draggin down that bar in tech, I guess.
Post reply on HN