Instructions to use Mitsua/mitsua-diffusion-one with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Mitsua/mitsua-diffusion-one with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Mitsua/mitsua-diffusion-one", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
Tiled images behave erratical during generation.
#2
by ShamelesslyAI - opened
Greetings! I tried to find the answer in google but no luck so far. I'm not sure if it is the model, or the UI, or something else (I'm running Mitsua Diffusion One from this notebook https://colab.research.google.com/drive/1ZqI5VmIyiPMD_nOI5ymvqcwAz2gM-bcF?usp=share_link).
Images are generating fine, but when I check "Tiling" all I get are random colors or shapes. Only once I managed to get a "readable" texture from the batch. So, I'm not sure if the setup I'm using is wrong or there is something with the MitsuaD1 model? (Tried SD-1.4 and worked fine).
Thank you!
Seems to be working fine now, something with the colab setup may have been the cause.
ShamelesslyAI changed discussion status to closed

