README.md

metadata

title: Veo 3.1 Fast API
emoji: 🎬
colorFrom: blue
colorTo: green
sdk: docker
app_file: app:app
app_port: 7860
hf_oauth: true
disable_embedding: true
hf_oauth_scopes:
  - inference-api

Veo 3.1 Fast - FastAPI Version

This is a FastAPI-based web application for generating videos from text or images using Hugging Face Inference Providers.

How to Use

1. Open the web interface.
2. Enter your Hugging Face User Access Token (it must start with hf_).
3. Choose either the "Text to Video" or "Image to Video" tab.
4. Fill in the required prompts and/or upload an image.
5. Click the generate button and wait for your video.

Note: This is a paid service that uses your personal Hugging Face credits for the fal-ai provider.

API Endpoints

• POST /text-to-video/: Requires prompt and token in a form data.
• POST /image-to-video/: Requires image (file), prompt, and token in a form data.