Instructions to use MintLab/FLUX-Krea-BLAZE with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use MintLab/FLUX-Krea-BLAZE with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("MintLab/FLUX-Krea-BLAZE", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Diffusion Single File
How to use MintLab/FLUX-Krea-BLAZE with Diffusion Single File:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- Draw Things
- DiffusionBee
it's worked very well
thanks for the great repo, it's worked very well with low vram (mine 3505 6GB) and absolutely fast (only took about less then 3 minutes for generate 1 image combine with lora+sage attention), and it's also can be combine with other flux dev loras.
my device
proc i5 gen 13
GPU RTX 3050 6GB
Ram 38GB
SSD 512GB
Works great in Forge for me.
OS: Win11
GPU: RTX 4080
Ram:64GB,
Works great in Forge for me.
OS: Win11
GPU: RTX 4080
Ram:64GB,
What settings are you using in Forge? I tried 8 steps with Uni-PC / SMG Uniform but the generated images are noise. I also tried Euler / Simple but I need to use more steps to get good results.
Works great in Forge for me.
OS: Win11
GPU: RTX 4080
Ram:64GB,What settings are you using in Forge? I tried 8 steps with Uni-PC / SMG Uniform but the generated images are noise. I also tried Euler / Simple but I need to use more steps to get good results.
try Euler A + SMG Uniform and Diffusion in Low Bits: bnb-fp4
I am unsure what is the best quality workflow. For me, this is same as default krea. I tried loras 32 64 128 i tried without loras and im bit confused.
Nah, it's worse than a Schnell LoRA