Gorse

Gorse is an open source recommender system engine written in Go. You import items, users and interaction data, and the system automatically trains models to generate recommendations for each user. It aims to be a universal recommender that can be added to a wide range of online services, such as content sites, shops and developer communities. Recommendations can come from several sources, including latest items, user to user and item to item similarity, and collaborative filtering. Gorse supports both classical recommenders and recommenders based on large language models, and it handles multimodal content such as text, images and video through embeddings. A graphical dashboard lets teams edit the recommendation pipeline and manage data, RESTful APIs cover data operations and recommendation requests, and online evaluation analyzes performance from recent user feedback. Storage options include Redis, MySQL, Postgres and MongoDB, and client libraries are shown for Go, Python, TypeScript, Java, Rust, Ruby, PHP and .NET. It is for engineers who want to run their own recommendation service. A playground mode starts with one Docker command and loads sample GitHub repository data for experimenting. The codebase is released under the Apache 2 license, so the software itself carries no license fee, and the vendor's site publishes no paid plans or hosted pricing.