Instructions to use blurgy/CoMPaSS-FLUX.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use blurgy/CoMPaSS-FLUX.1 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("black-forest-labs/FLUX.1-dev", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("blurgy/CoMPaSS-FLUX.1") prompt = "a photo of a laptop above a dog" image = pipe(prompt).images[0] - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
lora无法和flux fp8正常加载
lora key not loaded: double_blocks.0.processor.proj_lora1.down.weight
lora key not loaded: double_blocks.0.processor.proj_lora1.up.weight
lora key not loaded: double_blocks.0.processor.proj_lora2.down.weight
lora key not loaded: double_blocks.0.processor.proj_lora2.up.weight
lora key not loaded: double_blocks.0.processor.qkv_lora1.down.weight
A ComfyUI-compatible LoRA checkpoint has been uploaded at https://huggingface.co/blurgy/CoMPaSS-FLUX.1/blob/main/CoMPaSS-FLUX.1-comfyui.safetensors.
A ComfyUI custom node is implemented at https://github.com/blurgyy/CoMPaSS-FLUX.1-dev-ComfyUI.
Although loading this LoRA alone can already improve performance, we recommend using our custom node to also patch the FLUX model to obtain the best results.