LeanQuant's picture
Upload folder using huggingface_hub
bed57d0 verified
metadata
base_model:
  - black-forest-labs/FLUX.1-Krea-dev
base_model_relation: quantized
pipeline_tag: text-to-image
tags:
  - dfloat11
  - df11
  - lossless compression
  - 70% size, 100% accuracy

ComfyUI DFloat11 Model: black-forest-labs/FLUX.1-Krea-dev

  • 🚀 Fully Lossless – 100% identical outputs to the original model
  • 📦 32% smaller model size – lower VRAM requirement than float16
  • âš¡ Compatible with ComfyUI – drop-in support with custom nodes
  • 🔧 GPU-accelerated inference – optimized for CUDA 12.1+

Performance Comparison

Model Model Size Peak GPU Memory (1024×1024 image generation) Generation Time (A100 GPU)
FLUX.1-Krea-dev (BFloat16) 23.80 GB 24.28 GB 56 seconds
FLUX.1-Krea-dev (DFloat11) 16.33 GB 17.54 GB 58 seconds

Installing DFloat11 Nodes for ComfyUI

Please visit github.com/LeanModels/ComfyUI-DFloat11.

Workflow

Drag and drop the following image into ComfyUI.

Or download this JSON workflow and drop into ComfyUI.