Instructions to use ethanfel/Krea-2-Base-Diffusers with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use ethanfel/Krea-2-Base-Diffusers with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("ethanfel/Krea-2-Base-Diffusers", 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
Request for `convert_krea2_community_to_diffusers.py` conversion script
Hi, thank you for providing this working Diffusers conversion.
Could you share the scripts/convert_krea2_community_to_diffusers.py script mentioned in the model card, or point me to the repository where it is available?
I’m investigating the conversion path from Krea 2 community-format checkpoints to Krea2Transformer2DModel, particularly whether the same mapping could be applied to an FP8-scaled Turbo checkpoint.
Thanks!
Thank you very much for sharing the script! This was exactly what I was looking for.
I am investigating Krea 2 compatibility with a Diffusers-based application, and the script helped me clearly separate the original/community key conversion from the runtime support required by quantized checkpoints such as FP8.
The detailed key mapping and notes about last.up / last.down were especially helpful. I really appreciate the quick reply and the direct link!