You need to agree to share your Hugging Face information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

Simple Flux

Simple Flux offers a lightweight web interface to Flux using Gradio.

Hardware & software requirements

Tested on Ubuntu with NVIVIA RTX 4090 and GTX 1070 (8GB GPU RAM).

GPU Memory

If your GPU has less than 16GB GPU memory, you need to specify --low_mem when you start the script. Max GPU memory consumption that was observed when --low_mem was specified was 5.88GB, so 6GB GPU RAM may work.

System RAM

Tested on 64GB RAM hosts. I learned from a user that he/she was able to run it with 40GB RAM on Windows 11. It is unclear if you can run it on a machine with less RAM than that.

How to install Simple Flux

Currently, only systems with NVIDIA GPU and CUDA are supported. I am using Linux and these steps have not been tested on Windows yet.

Overview

Copy files from github
Set up virtual environment
Install Pytorch
Install other python packages

Steps

Copy files from github

Open the terminal and go to a directory that you want to install Simple Flux. Type:
```
git clone https://github.com/HowToSD/simple_flux.git
```
Set up virtual environment

Run the following. Note that if your system does not have python, you need to install it first.
```
python3 -m venv simpleflux
source simpleflux/bin/activate  # On Windows use `simpleflux\Scripts\activate`
```
Install Pytorch

ensure pip and setuptools are up to date:
```
pip install --upgrade pip setuptools
```
Install torch
```
pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121
```
For more information, refer to the installation section at https://pytorch.org/.
Install other python packages
```
pip install -r requirements.txt
```
Start Simple Flux Type:
```
python simple_flux.py
```
If you have low GPU memory, type:
```
python simple_flux.py --low_mem
```
This should start Gradio.
Using your browser, access the URL displayed in the above.

How to run simple_flux in Docker with nvidia GPU

install nvidia-container-toolkit

Build Docker image

docker build -t simple_flux_app:0.0.3 .

Run container (it takes long to download models, use later if you already have locally downloaded models):
```
docker run -p 7860:7860 --gpus all simple_flux_app:0.0.3
```
or, if you have huggingface models in cache (no need to download them anymore)
```
docker run -v ~/.cache/huggingface/hub:/root/.cache/huggingface/hub -p 7860:7860 --gpus all simple_flux_app:0.0.3
```
goto http://localhost:7860

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support