import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("Montey/dotnet-edge", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
edge-maxxing-newdream-sdxl
This holds the baseline for the SDXL Nvidia GeForce RTX 4090 contest, which can be forked freely and optimized
Some recommendations are as follows:
- Installing dependencies should be done in pyproject.toml, including git dependencies
- Compiled models should be included directly in the repository(rather than compiling during loading), loading time matters far more than file sizes
- Avoid changing
src/main.py, as that includes mostly protocol logic. Most changes should be inmodelsandsrc/pipeline.py
For testing, you need a docker container with pytorch and ubuntu 22.04.
You can download your listed dependencies with uv installed with:
pipx install uv
pipx ensurepath
You can then relock with uv lock, and then run with uv run start_inference
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support