You can deploy this model and start earning money today!

YGu1998/
SiD-DiT-Flux-1024-res

Name: SiD-DiT-Flux-1024-res
Availability: InStock

black-forest-labs/FLUX.1-dev

Views

No views yet

SiD-DiT: Score Distillation of Flow Matching Models

Model Card

SiD-DiT is a score-distillation pipeline for pretrained text-to-image flow-matching models. We introduce a general framework that can accelerate sampling in flow-matching models by distilling pretrained flow-matching teachers into high-quality few-step students, providing a principled way to unify acceleration techniques across both diffusion- and flow-based generators.

Installation

bash

# download model files from this repo
git clone https://huggingface.co/YGu1998/SiD-DiT-Flux-1024-res

# install packages

cd SiD_pipelines
pip install -r requirements.txt
cd ..

Inference with SiD-DiT

python

import torch
from SiD_pipelines import SiDFluxPipeline


if torch.cuda.is_available():
    torch_dtype = torch.bfloat16
    device = "cuda"
else:
    torch_dtype = torch.float32
    device = "cpu"

model_repo_id = "YGu1998/SiD-DiT-Flux-1024-res"

prompt = ["a studio portrait of an elderly woman smiling, soft window light, 85mm lens"]

pipe = SiDFluxPipeline.from_pretrained(
    model_repo_id,
    torch_dtype=torch_dtype,
).to(device)

generator = torch.Generator(device=device).manual_seed(42)


image = pipe(
    prompt=prompt,
    guidance_scale=1.0,
    num_inference_steps=4,
    width=1024,
    height=1024,
    generator=generator,
    time_scale=1000,
).images[0]

image.save("example.png")

License

Base model: black-forest-labs/FLUX.1-dev
Base license: FLUX.1 [dev] Non-Commercial License (flux-1-dev-non-commercial-license)

This checkpoint is a non-commercial derivative of FLUX.1 [dev] and is governed by the FLUX.1 [dev] Non-Commercial License. Under that license, the model and its derivatives may be used, modified, and distributed only for Non-Commercial Purposes, which include personal research and internal testing / non-commercial R&D in non-production environments. Revenue-generating or production use, or using this checkpoint to train or distill other models for commercial deployment, requires a separate commercial license from Black Forest Labs.

This repository does not grant any additional rights beyond the FLUX.1 [dev] license. You are responsible for ensuring that your use complies with the license and applicable law. Note that, per the base license, images generated by the model may generally be used for personal, scientific, or commercial purposes subject to the content and safety restrictions set by Black Forest Labs.