Instructions to use Tongyi-MAI/Z-Image-Turbo with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Tongyi-MAI/Z-Image-Turbo with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Tongyi-MAI/Z-Image-Turbo", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
Announcement: ComfyUI Flow Match Euler Discrete Scheduler
Hello team, thanks for this incredible model!
I made a custom node to reduce noise and get more details out of it in ComfyUI
try the Flow Match Euler Discrete Scheduler!
It's available here. https://github.com/erosDiffusion/ComfyUI-EulerDiscreteScheduler
Installed, restarted but does not show up in the comfyui ksampler
Wierd, it works for me. check the startup logs and see if there is any issue ? You can raise a bug if so.
Nothing in the startup logs. Using Nightly version. I am also on MAC if that has anything to do with it.
I found for me the import failed. Looking at the error, I needed a higher version of the python peft package. Had to upgrade to 0.17.0 (pip install peft==0.17.0 to upgrade it)
Interesting. The package should depend on diffusers. This must be an indirect dependency.
I am already on 0.17.0
Also tried 0.18.0 and the same. The scheduler does not show up. Not sure where I can see what failed
I think I found the error under the ComfyUI Manager.. Custom Nodes Manager.. Bringing up the node by searching name EulerFlowMatch and the error was in there. It's stupid you have to dig for it and it's not shown on the command line.
Nothing there. The EulerFlowMatch wont show in scheduler no matter what.
Well installing via git worked, not via comfyui manager.