Text Embedder

Lightweight Flask service that exposes a single endpoint to compute sentence embeddings using sentence-transformers (default model: sentence-transformers/all-MiniLM-L6-v2).

The repository includes a Dockerfile and compose.yml for easy containerized deployment and model caching.

Get started

docker compose -f compose.yml up --build

A prebuilt image is published to Github container registry on push to main branch.

Pull and run the image immediately:

docker pull ghcr.io/ablil/embedding-models-api:latest
docker run -p 8080:8080 ghcr.io/ablil/embedding-models-api:latest

API

Embedding

curl http://localhost:8080/embed -H 'content-type: application/json' -d '{"texts": ["first", "second"]}'

Health

curl http://localhost:8080/health