Oboyu

About

Oboyu is a semantic search system designed to create an intelligent index of text-based documents, with particular strength in processing Japanese language content. It indexes directories of text files, embeds their content using vector representations, and retrieves the most relevant documents based on user queries. Leveraging specialized Japanese tokenization and encoding detection, Oboyu excels in multilingual document collections, offering both a command-line interface and an MCP server mode for network-accessible document search.

Key Features

• Indexes local directories of text-based documents.
• Includes advanced reranking for improved search accuracy.
• Provides first-class support for Japanese text with specialized tokenization and encoding detection.
• 0 GitHub stars
• Offers semantic search using vector embeddings with the Ruri v3 model.
• Supports multiple search modes: vector, BM25, and hybrid.

Use Cases

• Indexing and searching a local collection of Japanese documents.
• Performing semantic search on multilingual text-based content.
• Building a privacy-focused, local document search engine.