Is there a good site that compares text-to-image models - showing a bunch of examples of text w/ output on each model?
FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs
41–50 of 159 posts
Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs
#42Earlier quoted context omitted.
Flux is more weird than old SD projects since Flux is extremely resource dependant and won't run on most hardware.
Doesn't take a lot of effort to get Flux dev/schnell to run on 3090s unquantized, but I agree that 24gb is the consumer GPU memory limit and there are many with less than that. Flux runs great on modern Mac hardware as well, if you have at least 32gb of unified memory.
I'm running Flux dev fine on a 3080 10GB, unquantised, on windows the nvidia drivers have a function to let it spill over into system ram. It runs a little slower, but it's not a deal-breaker unlike nvidia's pricing and power requirements at the moment
Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs
#43I asked for a simple scene and it drew in the exact same AI girl that every text-to-image model wants to draw, same face, same hair, so generic that a Google reverse image search pulls up thousands of the exact same AI girl. No variety of output at all.
Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs
#44Flux is so frustrating to me. Really good prompt adherence, strong ability to keep track of multiple parts of a scene, it's technically very impressive. However it seems to have had no training on art-art. I can't get it to generate even something that looks like Degas, for instance. And, I can't even fine tune a painterly art style of any sort into Flux dev. I get that there was working, living artist backlash at SD…
I’ve had the same problem with photography styles, even though the photographer I’m going for is Prokudin-Gorskii who used emulsion plates in the 1910s and the entire Library of Congress collection is in the public domain. I’m curious how they even managed to remove them from the training data since the entire LoC is such an easy dataset to access.
i'm fairly confident they did a broad FirstName LastName removal.
Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs
#45It doesn’t get piano keyboards right, but it’s the first image generator I’ve tried that sometimes get “someone playing accordion” mostly right. When I ask for a man playing accordion, it’s usually a somewhat flawed piano accordion, but If I ask for a woman playing accordion, it’s usually a button accordion. I’ve also seen a few that are half-button, half-piano monstrosities. Also, if I ask for “someone playing accor…
Periodic data is always hard for generative image systems - particularly if that "cycle" window is relatively large (as would be the case for octaves of a piano).
Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs
#46Are there any projects that allow for easy setup and hosting Flux locally? Similar to SD projects like InvokeAI or a1111
DrawThings on Mac
Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs
#47Flux is so frustrating to me. Really good prompt adherence, strong ability to keep track of multiple parts of a scene, it's technically very impressive. However it seems to have had no training on art-art. I can't get it to generate even something that looks like Degas, for instance. And, I can't even fine tune a painterly art style of any sort into Flux dev. I get that there was working, living artist backlash at SD…
I wonder if you can use Flux to generate the base image then img2img on SD1.4 to impart artistic style?
Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs
#48Pretty smart model. Here's one I made: https://replicate.com/p/6ez0x8xqvsrga0cjadg8m7bah0
Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs
#49Flux is so frustrating to me. Really good prompt adherence, strong ability to keep track of multiple parts of a scene, it's technically very impressive. However it seems to have had no training on art-art. I can't get it to generate even something that looks like Degas, for instance. And, I can't even fine tune a painterly art style of any sort into Flux dev. I get that there was working, living artist backlash at SD…
I've had a similar experience, incredible at generating a very specific style of image, but not great at generating anything with a specific style.
I suspect we'll see the answer to this is LoRAs. Two examples that stick out are:
- Flux Tarot v1 [0]
- Flux Amateur Photography [1]
Both of these do a great job of combining all the benefits of Flux with custom styles that seem to work quite well.
[0] https://huggingface.co/multimodalart/flux-tarot-v1 [1] https://civitai.com/models/652699?modelVersionId=756149
Re: FLUX1.1 [pro] – New SotA text-to-image model from Black Forest Labs
#50Flux is so frustrating to me. Really good prompt adherence, strong ability to keep track of multiple parts of a scene, it's technically very impressive. However it seems to have had no training on art-art. I can't get it to generate even something that looks like Degas, for instance. And, I can't even fine tune a painterly art style of any sort into Flux dev. I get that there was working, living artist backlash at SD…
I wonder if you can use Flux to generate the base image then img2img on SD1.4 to impart artistic style?
That's what a refiner is for in auto1111. Taking an image the last 10% and touching it up with an alternative model.
I actually use flux to generate image for purposes of adherence, then pull it in as a canny/depth controlnet with more established models like realvis, unstableXL, etc.