Online Quantization for Comfyui

Install

1. OneDiff Enterprise Installation Guide
2. OneDiffx Installation Guide
3. Before use, please confirm with python3 -m oneflow --doctor to confirm the argument enterprise: True. If the information displayed is as follows enterprise: True then it meets the requirements. If the information displayed is as follows enterprise: False, then run the following commandpip uninstall oneflow onediff_quant -y, and then follow the installation instructions for the Enterprise version to reinstall the OneDiff Enterprise version. You can find the relevant installation instructions through the following link: OneDiff Enterprise Installation Instructions

Usage:

Option	Range	Default	Description
quantized_conv_percentage	[0, 100]	100	Example value representing 100% quantization for linear layers
quantized_linear_percentage	[0, 100]	100	Example value representing 100% quantization for convolutional layers
conv_compute_density_threshold	[0, ∞)	100	Computational density threshold for quantizing convolutional modules to 100
linear_compute_density_threshold	[0, ∞)	300	Computational density threshold for quantizing linear modules to 300

Notes:

1. Specify the directory for saving graphs using export COMFYUI_ONEDIFF_SAVE_GRAPH_DIR="/path/to/save/graphs".
2. The log *.pt file is cached. Quantization result information can be found in cache_dir/quantization_stats.json.

Performance Comparison

SDXL

Accelerator	Baseline (non-optimized)	OneDiff(optimized)	OneDiff Quant(optimized)
NVIDIA GeForce RTX 3090	5.63 s	3.38 s ( ~40.0%)	2.60 s ( ~53.8%)

The following table shows the workflows used separately：

Note that you can download all images in this page and then drag or load them on ComfyUI to get the workflow embedded in the image.

Examples

Baseline (non-optimized)	OneDiff(optimized)	OneDiff Quant(optimized)

Download the required model files

For NA/EU users

cd ComfyUI
wget -O models/checkpoints/sd_xl_base_1.0.safetensors https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/resolve/main/sd_xl_base_1.0.safetensors

For CN users

cd ComfyUI
wget -O models/checkpoints/sd_xl_base_1.0.safetensors https://hf-mirror.com/stabilityai/stable-diffusion-xl-base-1.0/resolve/main/sd_xl_base_1.0.safetensors

Start ComfyUI

python main.py --gpu-only

SD1.5

Accelerator	Baseline (non-optimized)	OneDiff(optimized)	OneDiff Quant(optimized)
NVIDIA GeForce RTX 3090	3.54 s	2.13 s ( ~39.8%)	1.85 s ( ~47.7%)

The following table shows the workflows used separately：

Note that you can download all images in this page and then drag or load them on ComfyUI to get the workflow embedded in the image.

Examples

Baseline (non-optimized)	OneDiff(optimized)	OneDiff Quant(optimized)

Download the required model files

For NA/EU users

cd ComfyUI
wget -O  models/v1-5-pruned-emaonly.ckpt  https://huggingface.co/runwayml/stable-diffusion-v1-5/blob/main/v1-5-pruned-emaonly.ckpt

For CN users

cd ComfyUI
wget -O  models/v1-5-pruned-emaonly.ckpt  https://hf-mirror.com/runwayml/stable-diffusion-v1-5/resolve/main/v1-5-pruned-emaonly.ckpt

Start ComfyUI

python main.py --gpu-only

SVD

Accelerator	Baseline (non-optimized)	OneDiff(optimized)	OneDiff Quant(optimized)
NVIDIA A800-SXM4-80GB	35.54 s	25.59 s (27.99 %)	22.30 s (37.25 %)

The following table shows the workflows used separately：

Note that you can download all images in this page and then drag or load them on ComfyUI to get the workflow embedded in the image.

Examples

Baseline (non-optimized)	OneDiff(optimized)	OneDiff Quant(optimized)

Download the required model files

For NA/EU users

cd ComfyUI
wget -O  models/checkpoints/sd_xl_base_1.0.safetensors https://huggingface.co/stabilityai/stable-diffusion-xl-base-1.0/resolve/main/sd_xl_base_1.0.safetensors
wget -O  models/checkpoints/svd_xt_1_1.safetensors https://huggingface.co/vdo/stable-video-diffusion-img2vid-xt-1-1/resolve/main/svd_xt_1_1.safetensors

For CN users

cd ComfyUI
wget -O  models/checkpoints/sd_xl_base_1.0.safetensors https://hf-mirror.com/stabilityai/stable-diffusion-xl-base-1.0/resolve/main/sd_xl_base_1.0.safetensors
wget -O  models/checkpoints/svd_xt_1_1.safetensors https://hf-mirror.com/vdo/stable-video-diffusion-img2vid-xt-1-1/resolve/main/svd_xt_1_1.safetensors

Start ComfyUI

python main.py --gpu-only