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

Infinity

Infinity is an AI-oriented database for LLM applications, indexing and searching dense embeddings, sparse embeddings, tensors or multi-vectors, full text and structured data.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Infinity — 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 Infinity when an LLM application needs hybrid retrieval across several data representations rather than dense-vector search alone.

Who it is for

  • Developers building RAG and knowledge-search applications
  • Teams creating question-answering, conversational AI or copilot systems
  • Engineers needing hybrid retrieval for recommenders or semantic search

Documented uses

  • Combine semantic, sparse, tensor and full-text retrieval
  • Store and search records containing text, numeric fields and vectors
  • Provide a retrieval layer for RAG, recommendation and question-answering applications
  • Apply filtering and reranking across hybrid search results
Before you choose

Requirements & limits

The repository is public and uses the Apache-2.0 license. Documented deployment targets include Linux, macOS and Windows through WSL or WSL2, with Docker, binary deployment and source builds described.

  • Hybrid search combines dense embeddings, sparse embeddings, tensors or multi-vectors, full text and structured data.
  • Filtering and reranking are supported, including RRF, weighted sum and ColBERT.
  • The documented architecture can run as a single binary with no dependencies and can be embedded in Python as a module.
  • The documented Docker quickstart requires x86_64 with AVX2, Python 3.11+ and the relevant operating-system setup.
  • The project exposes Python and HTTP API documentation for application integration.

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

Deploy Infinity with Docker and run a vector search

Set up the Infinity server and Python client, then create a vector table and execute the documented dense-vector search example.

Before you start

  • x86_64 CPU with AVX2 support
  • Linux with glibc 2.17+, macOS, or Windows 10+ with WSL/WSL2
  • Python 3.11+
  • Docker for the documented container deployment

Steps

  1. Create the /var/infinity directory and pull the infiniflow/infinity:nightly Docker image.

  2. Start the Infinity container with the documented volume, file-descriptor limit and host-network settings.

  3. Install the Python client with pip install infinity-sdk==0.7.3.

  4. Connect to the server, create a table with integer, text and vector fields, insert the documented records, and apply dense matching with the Python API.

Expected outcome

A running Infinity server and a Python example that returns results from a dense-vector query.

Guide limitations

  • The documented deployment requires the stated CPU, operating-system and Python prerequisites.
  • The example uses a server address placeholder and demonstrates dense-vector matching rather than a complete application workflow.
  • Windows deployment requires WSL or WSL2; the documented Windows path also includes systemd, Docker Engine and host-networking steps.
Documentation 1 ↗
Evidence and transparency

Sources & review

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