can this fluxgym running on runpod?

[dekahobby]

is there any tutorial how to deploy it on runpod?

[engdante]

I installed fluxgym in runpod. However, in runpod I can't access 127.0.0.1

I also failed to make the script public. "To create a public link, set share=True in launch()."

Could you please explain more about this?

***** INSTALATION I am making a network drive that is mounted in /workspace

cd /workspace wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh && chmod +x Miniconda3-latest-Linux-x86_64.sh && ./Miniconda3-latest-Linux-x86_64.sh -p /workspace/miniconda3 -b

source /workspace/miniconda3/bin/activate conda create --prefix /workspace/fluxgym conda activate /workspace/fluxgym

git clone https://github.com/cocktailpeanut/fluxgym cd fluxgym git clone -b sd3 https://github.com/kohya-ss/sd-scripts

cd sd-scripts pip install -r requirements.txt

cd .. pip install -r requirements.txt

pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu121

pip install -U "huggingface_hub[cli]" huggingface-cli login --token {TOKEN!!!} --add-to-git-credential

huggingface-cli download comfyanonymous/flux_text_encoders clip_l.safetensors --local-dir /workspace/fluxgym/models/clip huggingface-cli download comfyanonymous/flux_text_encoders t5xxl_fp16.safetensors --local-dir /workspace/fluxgym/models/clip huggingface-cli download cocktailpeanut/xulf-dev ae.sft --local-dir /workspace/fluxgym/models/vae huggingface-cli download cocktailpeanut/xulf-dev flux1-dev.sft --local-dir /workspace/fluxgym/models/unet

[engdante]

**** [EDIT] -->> socat TCP-LISTEN:8288,fork TCP:127.0.0.1:7860 This works on port 8288 , but there is another problem.

When I upload the photos and create a dataset, the script does not save them in the "datasets" folder and then an error appears that such files do not exist.

[fahadshery]

**** [EDIT] -->> socat TCP-LISTEN:8288,fork TCP:127.0.0.1:7860 This works on port 8288 , but there is another problem.

I just created an issue #80 ... I am trying to dockerise this app as well but it's running on 127.0.0.1. were you able to change it to 0.0.0.0? it would make it easier for a reverse proxy interfacing with it in a container

[UmarMotion]

in line 1016 of app.py change to

demo.launch(server_name="0.0.0.0", server_port=7860, debug=True, show_error=True, allowed_paths=[cwd])

[fahadshery]

don't need to hard code the IP anymore. You can use the environment variable like export GRADIO_SERVER_NAME="0.0.0.0" and then run the app.py and it should work