/ant

Open-source vector database built to embedding similarity search

Primary LanguagePythonApache License 2.0Apache-2.0

Ant

python faiss_cpu license test

Ant is an open-source vector database built to embedding similarity search based on Faiss. Data backup draws on the hybrid mechanism of Redis's RDB and AOF to ensure data security and consistency.

Quick start

Build docker image.

$ docker build -t ant:latest .

Start ant server.

$ docker run ant:latest -p 1234:1234 -v {logs_dir}:/ant/logs {data_dir}:/ant/data

Restful API

Method URL description Status
POST /ant/create Create the instance by name.
POST /ant/delete Delete the instance by name.
GET /ant/list List all instances.
POST /ant/info Get the information of instance by name.
POST /ant/build Build the Faiss index of instance by input data.
POST /ant/insert Insert data to Faiss index of instance.
POST /ant/update Update the Faiss index of instance.
POST /ant/remove Remove data of the Faiss index of instance.
POST /ant/search K-nearest neighbor query from Faiss index.
GET /ant/bgsave Back up index data in the background.

Configuration

License

Apache-2.0