Text-to-Image
Diffusers
Safetensors
MageFlowPipeline
image-generation
image-editing
diffusion
rectified-flow
mage-flow
Instructions to use microsoft/Mage-Flow with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use microsoft/Mage-Flow with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("microsoft/Mage-Flow", torch_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 Settings
- Draw Things
- DiffusionBee
Censore
#5
by Czarj - opened
Read the code
Read the code
What do you mean? If you're referring to the CLI level issue, I've already found it. There were two validation blocks embedded in pipeline.py. Fortunately, Python is not compiled, making it easy to remove them. However, these Neanderthals hardcoded their flash attention into text_encoder and mage_layers, requiring me to spend a couple of hours fixing the hardcoding and making it work on Windows.
