Trim placeholder and model label to reflect direct inference

Add .gitignore for __pycache__ and .pyc

Remove __pycache__ from repo

Handle Gradio file references in to_data_url

Add return type annotation to chat() for streaming

Use client.submit() with for-await to stream data events

Redesign UI as terminal-style chat

Stream chat responses via SSE

Read HF_TOKEN from server-side Space secret

Revert to documented @app .get('/', response_class=HTMLResponse) pattern

Fix / route being intercepted by Gradio API

Convert to gradio.Server with custom HTML frontend

Upload folder using huggingface_hub

initial commit