OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Infrastructure

Antfly

Antfly is a Zig-based search-and-inference database combining BM25 full-text search, dense and sparse vector indexes, and graph traversal over shared tables. Its in-process models support chunking, embedding, reranking, transcription, OCR and extraction, while built-in agents support retrieval-augmented generation.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Antfly — OpenViST project profile in Infrastructure.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Antfly when an application needs multiple retrieval methods in one database, or when local embedding, inference and database-server deployment options are relevant.

Who it is for

  • Developers building hybrid search systems across keyword, semantic and graph retrieval.
  • Teams creating RAG applications that need indexing transformations and retrieval agents.
  • Developers embedding a local database or inference runtime in applications.

Documented uses

  • Index data into full-text, dense-vector and sparse-vector search structures.
  • Combine retrieval results and graph traversal over shared tables.
  • Generate chunks, embeddings, entities and graph edges as data is written.
  • Build RAG interactions with streaming, multi-turn chat and tool calling.
  • Index and search images, audio and video.
Before you choose

Requirements & limits

The documented CLI quickstarts target macOS and Linux, with Homebrew and Docker alternatives. The repository identifies Antfly Lite and inference components as Apache 2.0 and the full database server as ELv2. Local-development quickstarts run without authentication; remote access requires enabling authentication, setting an admin password and configuring TLS at a reverse proxy.

  • Choose between embedded Lite, standalone node, multi-Raft cluster or serverless over object storage according to deployment needs.
  • The full-text, dense-vector, sparse-vector and graph capabilities share tables and can be combined for retrieval.
  • The repository distinguishes Lite and inference licensing from the full server's ELv2 license.
  • Local quickstarts are unauthenticated and intended for development; remote operation requires the documented security setup.
  • Antfly's inference runtime includes tasks such as chunking, embedding, reranking, OCR and transcription.

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

Build the Antfly Lite CLI and initialize a database file

Compile the Lite CLI and create an `.aflite` database file from the repository source.

Before you start

  • Antfly source checkout
  • Zig build toolchain

Steps

  1. Change to the zig directory.

  2. Run zig build lite -j1.

  3. Run ./zig-out/bin/antfly-lite lite init app.aflite.

Expected outcome

The Lite build provides the CLI, native library and C header, and the command initializes `app.aflite`.

Guide limitations

  • This procedure initializes a Lite database file; serving it over HTTP requires the full database server executable.
  • The Lite build and library are identified as Apache 2.0, while the full database server is identified as ELv2.
Documentation 1 ↗
Evidence and transparency

Sources & review

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