🌍 Multilingual Embedding API

A FastAPI-powered service for generating multilingual sentence embeddings using the intfloat/multilingual-e5-large model. Supports local inference, token-based authentication via API keys, and can be deployed easily using Docker.

---

🚀 Features

• Supports 100+ languages
• Local embedding generation (no external API calls)
• Cosine similarity-ready embeddings
• API key authentication
• Dockerized for easy deployment

---

📦 Setup

1. Clone the Repository

git clone https://github.com/Kitisaks/embeddy.git
cd embeddy

2. Generate an API Key

add it to .env:

EMBED_API_TOKEN=yourtokensecret

3. Build & Run with Docker

docker build -t embeddy .
docker run -p 8000:8000 embeddy

📚 References

• SentenceTransformers

• FastAPI Docs

• Hugging Face Model