VisionForge AI

This project is a simple full-stack prototype for generating cinematic scene concepts, storyboards, shot lists, and images.

✅ What works now

• Backend API server (FastAPI) that returns director analysis, storyboard, and shot list.
• Frontend (React + Vite) that calls the backend and displays results.
• Image generation falls back to a placeholder image when Stable Diffusion is not available.

🚀 Running locally (recommended)

1) Start the backend

1. Open a terminal at visionforge-ai/backend
2. (Optional) Create a virtual environment:

   python -m venv .venv
   .\.venv\Scripts\Activate.ps1

3. Install dependencies:

   pip install -r requirements.txt

4. Start the backend server:

   python main.py

The backend will be available at: http://127.0.0.1:8000

2) Start the frontend

1. Open a new terminal at visionforge-ai/frontend
2. Install dependencies:

   npm install

3. Start the dev server:

   npm run dev

The frontend will open in your browser at http://localhost:5173 and proxy requests to the backend.

🔧 Optional: Enable real OpenAI-style LLM + Stable Diffusion image generation

LLM (Groq)

1. Add a GROQ_API_KEY to backend/.env.
2. Optionally change GROQ_MODEL in backend/.env.

Stable Diffusion (local)

1. Set ENABLE_LOCAL_SD=true in backend/.env.
2. Install the required packages (can be large):

   pip install diffusers[torch] torch

Stable Diffusion (remote via Hugging Face)

1. Add HUGGINGFACE_API_TOKEN=<your_token> to backend/.env (go to https://huggingface.co/settings/tokens to create one).
2. Optionally set HF_MODEL to a model you have access to (defaults to stabilityai/stable-diffusion-xl-base-1.0).

OpenAI Images (remote)

1. Add OPENAI_API_KEY=<your_key> to backend/.env.
2. Optionally set OPENAI_IMAGE_MODEL=gpt-image-1 in backend/.env.
3. The app will use the OpenAI Images generations endpoint to generate a landscape image.

If no image backend is available, the app will serve a placeholder image and continue working.