Skip to content

README.md

Latest commit

 

History

History
19 lines (15 loc) · 774 Bytes

README.md

File metadata and controls

19 lines (15 loc) · 774 Bytes

host_video_generation

This repository contains a minimal FastAPI application exposing a dummy video generation service. It mimics the request/response structure of vLLM's OpenAI compatible API but returns placeholder video files instead of text.

Endpoints

  • POST /generate – Accepts a JSON body with prompt, base64 encoded image, optional negative_prompt and guidance fields. Returns the path to a generated video file.
  • GET /download/{file_id} – Download a previously generated dummy MP4 file.
  • GET /ping – Basic health‑check that also reports environment configuration.

The service does not run any actual video generation pipeline. It is intended for development and integration tests.