Fix a syntactic bug.

Fix syntax bug.

Display the notification only when the model is changed.

Update the deps.

Fix the warning when the model fails to load.

Add the sampling metric + update the docs.

Update / refactor the app, with the test samples.

Download the models without instantiating them.

Warn the user when the calculations are aborted.

Allow to switch between a range of models.

Update psaiops to manage several models.

Increase the GPU allocation for the 27b model.

Disable the debugging mode.

Update psaiops.

Update the deps.

Remove useless arguments.

Change the title and default models.

Upgrade psaiops (fixes) + relax other deps.

Use Gemma 3 instead of GPT-OSS.

Update the deps, which solve the issues with `GptOssForCausalLM`.

Remove unrelated code + use the GPU wrapper on the main process.

Rebrand the app as a human / LLM detector.

Update the deps.

Cannot access the GPU outside of the spaces wrapper.

Load custom CSS styles to give a visual feedback on button click.

Start from another app + use the latest API

initial commit