padaniga

[the stable audio open small memo thang]

sudo apt install python3 python3-pip

python3 -m venv diffenv

source diffenv/bin/activate

pip install diffusers transformers torch accelerate

For CLI tools like huggingface-cli, you can install them globally but cleanly with pipx:

sudo apt install pipx

pipx ensurepath

pipx install huggingface_hub

pipx upgrade huggingface_hub

pipx upgrade-all

pip install torch torchaudio transformers accelerate

pip install --upgrade diffusers transformers torch accelerate

pip install diffusers transformers accelerate

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

pip list --outdated

# Then upgrade all manually or run:

pip list --outdated | tail -n +3 | awk '{print $1}' | xargs -n1 pip install -U

pip install --upgrade pip

pip install huggingface_hub

huggingface-cli login

https://huggingface.co/stabilityai/stable-audio-open-1.0

https://huggingface.co/settings/tokens

git clone https://huggingface.co/stabilityai/stable-audio-open-1.0

~/.huggingface/token

from_pretrained("model-name", token="your_token_here")

export HUGGINGFACE_HUB_TOKEN=your_token_here

deactivate

git clone https://github.com/stability-ai/stable-audio-open-mini.git

cd stable-audio-open-mini

pip install -e .

pip install -r requirements.txt

Once installed, you can delete the .git folder if you don't need to pull updates:

rm -rf .git

pipe = StableAudioOpenPipeline.from_pretrained("stabilityai/stable-audio-open-small")

https://chatgpt.com/share/687b06d3-bc8c-8010-a70b-d27a7496e42f