DAINAPP
patreon
Stable Diffusion 0.2
🕑 Added 2022-08-27 19:03:50 +0000 UTCFor this update I focused more on two of the most requested changes.
img2img: Select an image as a starting point to generate Images
Working Samples: Generate more than 1 image per render, this also use more Vram.
Other than that I did a little optimization and hide some unused options.
There still looooots of stuff to do, so expect more updates.
Unrelated note:
My new card is here, but I also had to upgrade my motherboard to use the 3080. Surprise, my new motherboard came is with a hardware problem 😂 So now i'm talking to the seller, it's gonna be easy to change it, but it will take some extra time to start using the new card, well, at least it don't stop my work, just maybe make it a little slower.
Comments
GRisk
Hi Gutirz, can you try adding the folder to the /models folder? It should work similar to what you are doing
Any chance to still get this version? I am using dreambooth in colab, it works by exporting the concept folder from colab and replacing files in the /diffuser16 folder inside /Stable Diffusion GRisk GUI. I can do it on the 0.1 version, but was wondering if there was any newer version that I could use - the most recent ones don't have the folder.
Greeneye
I'm only able to get 576 x 576 at the most out of my 3080. And when I do that it runs much slower. I tend to just stay at 512x512 or less
Is it possible to get 1024px by 1024px images on 16gb rtx 3080? I get the message and can't understand what should I do with it: "RuntimeError: CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 16.00 GiB total capacity; 9.93 GiB already allocated; 3.58 GiB free; 10.03 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and P"
