Releases·Tool Release
Xtriever: on-device RAG retrieval library ships for Rust, Python, Swift, and Kotlin
For anyone doing local (or not) RAG: I built [Xtriever](https://github. com/mirth/xtriever), the retrieval half of the problem, designed to run on the device with no server
0 engagement1 sourceFri, Sep 25, 2026, 03:09 PM
AI Overview
- Developer mirth releases Xtriever, a serverless retrieval library for local RAG, with bindings for Rust, Python, Swift, and Kotlin.
- Xtriever offers demo apps for iOS and Android plus a Python CLI; it is not production ready and must be built from source.
- Xtriever lacks a learned ranker, and its Android benchmarks come only from emulator measurements; the project is, per the author, "purely vibecoded".
Related
Releases·Tool Release
raggy - CLI tool for RAG over your local documents built with Ollama
Developer paulknysh released raggy, a lightweight CLI tool for retrieval-augmented generation over local documents, built on Ollama. It runs vector search, BM25 indexing, and embedding generation entirely locally, with answer generation via a local LLM or a remote API key. The tool supports common document formats and applies OCR to images and scans automatically.
Developer paulknysh released raggy, a lightweight CLI tool for retrieval-augmented generation over local documents, built on Ollama. It runs vector search, BM25 indexing, and embedding generation entirely locally, with answer generation via a local LLM or a remote API key. The tool supports common document formats and applies OCR to images and scans automatically.
0 engagement1 sourceredditWed, Sep 9, 2026, 03:57 PM
Products·launched
CORTEX RAG - self-hosted RAG pipeline with exposed retrieval components
CORTEX RAG is an open-source RAG system that runs locally and keeps documents on the user's own infrastructure, exposing the retrieval pipeline rather than hiding it behind a hosted API. LLMs power retrieval techniques like contextual retrieval, GraphRAG, corrective RAG, HyDE, and neural reranking to produce source-aware answers. It targets developers and teams who want full control over their document data instead of chat-with-PDF demos.
CORTEX RAG is an open-source RAG system that runs locally and keeps documents on the user's own infrastructure, exposing the retrieval pipeline rather than hiding it behind a hosted API. LLMs power retrieval techniques like contextual retrieval, GraphRAG, corrective RAG, HyDE, and neural reranking to produce source-aware answers. It targets developers and teams who want full control over their document data instead of chat-with-PDF demos.
4 engagement1 sourceredditFri, Sep 25, 2026, 03:55 AM
Products·launched
Raggy – lightweight CLI tool for RAG over local documents
Raggy is an open-source command-line tool for running retrieval-augmented generation over documents stored locally. It uses LLMs to answer questions grounded in the user's own files without uploading them to external services. It targets developers who want a simple, self-hosted way to search and query local document collections.
Raggy is an open-source command-line tool for running retrieval-augmented generation over documents stored locally. It uses LLMs to answer questions grounded in the user's own files without uploading them to external services. It targets developers who want a simple, self-hosted way to search and query local document collections.
7 engagement1 sourcehackernewsFri, Sep 11, 2026, 04:01 PM
Products·building
Local AI memory layer for personal file retrieval
A fully local memory and retrieval engine that indexes personal files (PDFs, notes, screenshots, code) using embeddings and vector search. It solves the problem of finding information in personal files when the user cannot recall the filename or exact wording. The system runs entirely on-device, prioritizing privacy and offline capability.
A fully local memory and retrieval engine that indexes personal files (PDFs, notes, screenshots, code) using embeddings and vector search. It solves the problem of finding information in personal files when the user cannot recall the filename or exact wording. The system runs entirely on-device, prioritizing privacy and offline capability.
2 engagement1 sourceredditWed, Jul 15, 2026, 05:24 PM
Products·launched
Local Search Agent: offline RAG with no embeddings
A local search agent that performs retrieval-augmented generation (RAG) entirely offline without using embeddings. It solves the problem of running AI-powered search on private documents without sending data to external services, targeting privacy-conscious users or those with limited internet access.
A local search agent that performs retrieval-augmented generation (RAG) entirely offline without using embeddings. It solves the problem of running AI-powered search on private documents without sending data to external services, targeting privacy-conscious users or those with limited internet access.
1 engagement1 sourcehackernewsMon, Jul 13, 2026, 06:49 PM