OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
RAG & Knowledge Systems

Rustyrag

RustyRAG is an open-source Rust API for building document question-answering systems. It indexes extracted content in Milvus, combines HNSW vector search with BM25 keyword retrieval, optionally reranks passages, and streams LLM-generated answers with source citations. Its documented local stack includes Actix-web, Docker Compose, Docling, Jina embeddings, Groq and Cerebras integrations.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Rustyrag — OpenViST project profile in RAG & Knowledge Systems.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use RustyRAG when you need to build and operate a document-grounded Q&A backend and can configure its Rust, Docker, database, extraction, model-provider and embedding services.

Who it is for

  • Developers building a document-grounded API for an application or assistant.
  • Teams creating searchable knowledge bases from PDFs, office files, web pages or text.
  • Researchers and technical users who want to experiment with hybrid retrieval and reranking.

Documented uses

  • Ingest supported documents and extract their text and layout for indexing.
  • Retrieve relevant passages using both semantic vector search and keyword matching.
  • Rerank retrieved passages and return a streamed answer with citations.
  • Operate a local RAG backend with interactive chat and Swagger API documentation.
Before you choose

Requirements & limits

The repository describes RustyRAG as open source and provides a self-hosted quick-start. The documented setup requires Groq and Cerebras API keys.

  • Rust 1.70 or later and Docker Compose are listed as setup prerequisites.
  • The local stack includes Milvus, Docling and Jina embedding and reranking services; embeddings, reranking and Docling use GPU, while the Rust server runs on CPU.
  • Reranking improves the README’s reported pass rate from 91.6% to 94.5%, while increasing average time to first token from 181 ms to 279 ms in its benchmark setup.
  • The README lists TXT, PDF, DOCX, PPTX, XLSX, HTML and ZIP input.
  • The benchmark figures reflect a specified Open RAG Bench test setup and should not be treated as performance guarantees.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up RustyRAG locally

Build and start the documented local RAG API and open its chat or API documentation interfaces.

Before you start

  • Rust 1.70 or later
  • Docker and Docker Compose
  • Groq API key and Cerebras API key

Steps

  1. Clone the repository and enter its directory; copy .env.example to .env.

  2. Set GROQ_API_KEY and CEREBRAS_API_KEY in .env.

  3. Run the command below to start Docling, Jina embeddings, the reranker and Milvus.

    Documented command
    docker compose up -d

  4. Build and start the server with the command below and the command below.

    Documented command
    cargo build --release
    Documented command
    cargo run --release

  5. Open the local chat interface at http://localhost:8080/static/chat.html or Swagger UI at http://localhost:8080/swagger-ui/.

Expected outcome

A local RustyRAG server with the documented chat and API documentation interfaces available; the README describes uploading a PDF and asking a question to receive streamed answers with citations.

Guide limitations

  • The guide documents local setup and interfaces, not production deployment or a successful execution test.
  • The README’s clone command uses https://github.com/AlphaCorp-AI/RustyRAG, while the supplied project repository is https://github.com/RustyRAG/RustyRAG; confirm the intended clone URL before using the command.
Documentation 1 ↗
COMPARE OPTIONS

Compare options for Vector search

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
RustyragIngest supported documents and extract their text and layout for indexing.Self-hosted, ServerThe repository describes RustyRAG as open source and provides a self-hosted quick-start. The documented setup requires Groq and Cerebras API keys.
MettisIndex text documents through an HTTP endpoint.Linux, Self-hosted, Server
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
16 October 2026
Original website title
RustyRAG/RustyRAG · Official homepage title · retained in its source language
How we review sourcesSuggest a correction