Svdq Int4 Flux.1 Depth Dev

🚀 svdq-int4-flux.1-depth-dev

svdq-int4-flux.1-depth-dev is an INT4-quantized version of the FLUX.1-Depth-dev model. It can generate images based on text descriptions while following the structure of the input image. This quantized model offers about 4× memory savings and runs 2 - 3× faster than the original BF16 model.

✨ Features

• Quantization: Utilizes INT4 quantization to reduce memory usage and accelerate inference.
• Image Generation: Capable of generating images from text descriptions while adhering to the input image's structure.
• Low - Rank Decomposition: SVDQuant method decomposes weights into low - rank components and residuals for easier quantization.
• Optimized Engine: Nunchaku engine optimizes the low - rank branch to reduce latency overhead.

📦 Installation

Diffusers

Please follow the instructions in [mit - han - lab/nunchaku](https://github.com/mit - han - lab/nunchaku) to set up the environment. Also, install some ControlNet dependencies:

pip install git+https://github.com/asomoza/image_gen_aux.git
pip install controlnet_aux mediapipe

💻 Usage Examples

Basic Usage

import torch
from diffusers import FluxControlPipeline
from diffusers.utils import load_image
from image_gen_aux import DepthPreprocessor

from nunchaku.models.transformer_flux import NunchakuFluxTransformer2dModel

transformer = NunchakuFluxTransformer2dModel.from_pretrained("mit-han-lab/svdq-int4-flux.1-depth-dev")

pipe = FluxControlPipeline.from_pretrained(
    "black-forest-labs/FLUX.1-Depth-dev",
    transformer=transformer,
    torch_dtype=torch.bfloat16,
).to("cuda")

prompt = "A robot made of exotic candies and chocolates of different kinds. The background is filled with confetti and celebratory gifts."
control_image = load_image("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/robot.png")

processor = DepthPreprocessor.from_pretrained("LiheYoung/depth-anything-large-hf")
control_image = processor(control_image)[0].convert("RGB")

image = pipe(
    prompt=prompt, control_image=control_image, height=1024, width=1024, num_inference_steps=30, guidance_scale=10.0
).images[0]
image.save("flux.1-depth-dev.png")

Comfy UI

Work in progress. Stay tuned!

📚 Documentation

Method

Quantization Method -- SVDQuant

Overview of SVDQuant. Stage1: Originally, both the activation X and weights W contain outliers, making 4 - bit quantization challenging. Stage 2: We migrate the outliers from activations to weights, resulting in the updated activation and weight. While the activation becomes easier to quantize, the weight now becomes more difficult. Stage 3: SVDQuant further decomposes the weight into a low - rank component and a residual with SVD. Thus, the quantization difficulty is alleviated by the low - rank branch, which runs at 16 - bit precision.

Nunchaku Engine Design

(a) Naïvely running low - rank branch with rank 32 will introduce 57% latency overhead due to extra read of 16 - bit inputs in Down Projection and extra write of 16 - bit outputs in Up Projection. Nunchaku optimizes this overhead with kernel fusion. (b) Down Projection and Quantize kernels use the same input, while Up Projection and 4 - Bit Compute kernels share the same output. To reduce data movement overhead, we fuse the first two and the latter two kernels together.

Model Description

🔧 Technical Details

• The model is based on the FLUX.1 - Depth - dev model and is quantized to INT4.
• It uses the SVDQuant method for quantization and the Nunchaku engine for inference optimization.

📄 License

The model is licensed under the Apache - 2.0 license.

Citation

If you find this model useful or relevant to your research, please cite

@inproceedings{
  li2024svdquant,
  title={SVDQuant: Absorbing Outliers by Low-Rank Components for 4-Bit Diffusion Models},
  author={Li*, Muyang and Lin*, Yujun and Zhang*, Zhekai and Cai, Tianle and Li, Xiuyu and Guo, Junxian and Xie, Enze and Meng, Chenlin and Zhu, Jun-Yan and Han, Song},
  booktitle={The Thirteenth International Conference on Learning Representations},
  year={2025}
}

Quantization Library: DeepCompressor   Inference Engine: Nunchaku

⚠️ Important Note

• The model is only runnable on NVIDIA GPUs with architectures sm_86 (Ampere: RTX 3090, A6000), sm_89 (Ada: RTX 4090), and sm_80 (A100). See this issue for more details.
• You may observe some slight differences from the BF16 models in detail.