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

41–50 of 237 posts

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

#41

Post by orra »

rgbrgb wrote: post #30
ilaksh wrote: post #27

Earlier quoted context omitted.

They are technically open source. It's just that the model license prohibits commercial use and the code license prohibits bypassing the filters. So it's kind of worse than closed source in a way because it's like a tease. With no API apparently. Theoretically large companies or rich people might be able to make a licensing agreement.

> model license prohibits commercial use I thought that at first, but I think it only prohibits commercial use that breaks regional copyright or privacy laws.

It prohibits both commercial use, whether or not you break regional laws; and it prohibits breaking certain laws. As another user said, encoding the law into a licence is pointless but makes it non-free.

There are also problematic restrictions on your ability to modify the software under clause 2(c). And nor do you have the right to sublicence, it's not clear to me what rights somebody has if you give them a copy.

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

#42

Post by jacob019 »

Any web based front ends yet? I put together a system that runs a variety of web based open source AI image generation and editing tools on Vultr GPU instances. It spins up instances on demand, mounts an NFS filesystem with local caching and a COW layer, spawns the services, proxies the requests, and then spins down idle instances when I'm done. Would love to add this, suppose I could whip something up if none exists.

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

#43

Post by orra »

connerruhl wrote: post #40
orra wrote: post #11

Neither the source code nor the weights are open source... This is actually worse than Stability AI's previous offering, in that regard.

That'll change when the full non-research release occurs... https://twitter.com/EMostaque/status/1651328161148174337

That tweet is vague. Besides, it says like 'like SD', so I will be pleasantly shocked if the models are open source.

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

#45

Post by Jackson__ »

oh_sigh wrote: post #26
hunkins wrote: post #19

New restriction in their License suggests the software can't be modified. "2. All persons obtaining a copy or substantial portion of the Software, a modified version of the Software (or substantial portion thereof), or a derivative work based upon this Software (or substantial portion thereof) must not delete, remove, disable, diminish, or circumvent any inference filters or inference filter mechanisms in the Softwar…

You can't remove the filters per the license, but the weights will be available soon and so anyone can just reimplement this code using the weights

There is a similar license clause for the weights[0] as well, so I'm not sure this would apply unless you write the code and train your model from scratch.

[0] https://github.com/deep-floyd/IF/blob/main/LICENSE-MODEL#L54

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

#46

Post by CamperBob2 »

coolspot wrote: post #35
TaylorAlexander wrote: post #6

Earlier quoted context omitted.

If you don't mind the power consumption I noticed that older nvidia P6000's (24GB) are pretty cheap on ebay! My 16GB P5000 is pretty handy for this stuff.

Looks like P6000 24Gb goes for $800-$1200 while you can get superior 3090 24Gb for $800-$1000 .

4090s are only $1600 or so now, for that matter.

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

#47

Post by tmabraham »

Taek wrote: post #29

For anyone who doesn't know, DeepFloyd is a StableDiffusion style image model that more or less replaced CLIP with a full LLM (11b params). The result is that it is much better at responding to more complex prompts. In theory, it is also smarter at learning from its training data.

It isn't like Stable Diffusion, it's more like Google's Imagen model.

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

#48

Post by TaylorAlexander »

coolspot wrote: post #35
TaylorAlexander wrote: post #6

Earlier quoted context omitted.

If you don't mind the power consumption I noticed that older nvidia P6000's (24GB) are pretty cheap on ebay! My 16GB P5000 is pretty handy for this stuff.

Looks like P6000 24Gb goes for $800-$1200 while you can get superior 3090 24Gb for $800-$1000 .

oh! My mistake thanks for letting me know.

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

#50

Post by ronsor »

oh_sigh wrote: post #26
hunkins wrote: post #19

New restriction in their License suggests the software can't be modified. "2. All persons obtaining a copy or substantial portion of the Software, a modified version of the Software (or substantial portion thereof), or a derivative work based upon this Software (or substantial portion thereof) must not delete, remove, disable, diminish, or circumvent any inference filters or inference filter mechanisms in the Softwar…

You can't remove the filters per the license, but the weights will be available soon and so anyone can just reimplement this code using the weights

That's already been done: https://github.com/lucidrains/imagen-pytorch
Post reply on HN