DeepFloyd IF: open-source text-to-image model
1–10 of 237 posts
Re: DeepFloyd IF: open-source text-to-image model
#2"Imagen free"
Re: DeepFloyd IF: open-source text-to-image model
#3Currently down on hugging face
Re: DeepFloyd IF: open-source text-to-image model
#416GB VRAM minimum is a bit steep. Sadly excludes my 3080 which is annoying because I'd like something better than Stable Diffusion locally.
Re: DeepFloyd IF: open-source text-to-image model
#5Has anyone tried the Scott Alexander AI bet prompts?
1. A stained glass picture of a woman in a library with a raven on her shoulder with a key in its mouth
2. An oil painting of a man in a factory looking at a cat wearing a top hat
3. A digital art picture of a child riding a llama with a bell on its tail through a desert
4. A 3D render of an astronaut in space holding a fox wearing lipstick
5. Pixel art of a farmer in a cathedral holding a red basketball
Re: DeepFloyd IF: open-source text-to-image model
#616GB VRAM minimum is a bit steep. Sadly excludes my 3080 which is annoying because I'd like something better than Stable Diffusion locally.
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.
Re: DeepFloyd IF: open-source text-to-image model
#7So this one can create perfect text in images? If true, that’s insane
Re: DeepFloyd IF: open-source text-to-image model
#8interesting there are different models: https://github.com/deep-floyd/IF#-model-zoo-
I'm also very happy for the release of the two upscaler, I can use them to upscale to result of my small 64x64 DDIM models (maybe with some finetuning).
Re: DeepFloyd IF: open-source text-to-image model
#916GB VRAM minimum is a bit steep. Sadly excludes my 3080 which is annoying because I'd like something better than Stable Diffusion locally.
There's a note which suggests you might be able to get by on lower. My 3060 struggles with SD on the defaults, but works fine with float16.
There are multiple ways to speed up the inference time and lower the memory consumption even more with diffusers. To do so, please have a look at the Diffusers docs:
Optimizing for inference time [1]
Optimizing for low memory during inference [2]
[2] https://huggingface.co/docs/diffusers/api/pipelines/if#optim...
Re: DeepFloyd IF: open-source text-to-image model
#10So this one can create perfect text in images? If true, that’s insane
LDM-400M was already able to generate text (predecessor of Stable Diffusion), thanks to the fact that every token in the text encoder (trained from scratch) was available in the attention layer.