README.md

---
title: t5-smaller Demo
emoji: 🤏
colorFrom: indigo
colorTo: blue
sdk: gradio
sdk_version: 5.49.1
python_version: 3.12.12
app_file: app.py
models:
  - ShinpacheShimura/t5-smaller
---

# t5-smaller Space

A small Gradio demo for
[`ShinpacheShimura/t5-smaller`](https://huggingface.co/ShinpacheShimura/t5-smaller).

The checkpoint currently keeps its model files in
`optimized-flan-t5-small`, so `app.py` passes that directory as `subfolder`.

## Deploy

1. On Hugging Face, select **New Space**.
2. Choose **Gradio** as the SDK.
3. Upload `README.md`, `app.py`, and `requirements.txt` from this folder.
4. Open **Settings → Hardware** and select **Nvidia T4 – small** or another GPU.
5. Wait for the Space to rebuild.

The demo will appear when the build finishes. Its **Use via API** link provides
Python and JavaScript client examples automatically.

This Space uses lazy loading: the model is downloaded and loaded on the first
request, not while the UI is being constructed.