Deployment

Frontend Deployment (Vercel)

Build Command:

npm run build

Output Directory:

/client/dist

Required Environment Variable:

VITE_API_BASE_URL=https://yourbackend.render.com

Backend Deployment (Render/AWS)

Start Command:

uvicorn main:app --host 0.0.0.0 --port 8080

GPU Requirements:

NVIDIA GPU with CUDA support
Minimum 8GB VRAM (16GB recommended for batch processing)
CUDA 11.8 or higher

Recommended Instance Types:

AWS: g4dn.xlarge or larger
Render: GPU instances with at least 8GB VRAM

Need help? Contact support@elitetranscript.ai