Show HN: Xtriever – offline RAG retrieval on a phone

Category: ai-ml

Tags: rag, on-device, hybrid-search, rust, mobile, retrieval

Score: 7.5/10 (Innovation: 7, Technical: 9, Documentation: 8, Utility: 6)

Xtriever is an offline hybrid retrieval engine for RAG, combining BM25, dense embeddings, fusion, and cross-encoder re-ranking, all running on-device (mobile/laptop) in Rust. It offers native bindings for Python, Swift, and Kotlin, with strong determinism and verification guarantees. The project is technically sophisticated, demonstrating a complete multi-stage retrieval pipeline with performance measurements and quality benchmarks.

Target audience: mobile developers, backend developers, ai engineers

Repository: https://github.com/mirth/xtriever · Rust · Apache-2.0 · 4 stars

View on Hacker News