Omnigraph

Omnigraph provides a graph engine for lakehouse data with Git-style workflows.

github.com/ModernRelay/omnigraph
Omnigraph homepage screenshot

About Omnigraph

Omnigraph is an open-source, lakehouse-native graph engine developed by ModernRelay and hosted on GitHub. It is designed to serve as an operational state and coordination layer for fleets of AI agents, with built-in Git-style branching and merging workflows for graph data management.

Users can deploy Omnigraph as a server, declare graph configurations, schemas, stored queries, embedding providers and policies as code via a cluster.yaml file, and run it on their own infrastructure using local storage or any S3-compatible object store. It supports multimodal retrieval combining graph traversal, vector approximate nearest neighbor search, full-text search and Reciprocal Rank Fusion in a single query runtime, and enforces server-side Cedar policies for all data mutations. Hundreds of agents can work on parallel isolated graph branches, with changes reviewed and merged safely.

The tool is suited for teams building agentic systems, organizational knowledge graphs, agent memory stores, R&D/ML data layers, or development graphs for coding agents, with pre-built cookbooks available for common domain-specific use cases.

Key features

  • Git-style branching and merging for graph data changes
  • Built for fleets of parallel AI agents
  • Multimodal retrieval combining graph, vector and full-text search
  • Declarative cluster configuration via cluster.yaml files
  • Server-side Cedar policy enforcement for all mutations
  • Runs on local storage or S3-compatible object stores
Best for: Teams building fleets of AI agents that require versioned, auditable graph data for knowledge management, context assembly or R&D use cases.
Good for
  • Open-source with publicly available source code
  • Runs on user-controlled infrastructure, no external data hosting
  • Supports parallel agent work on isolated graph branches
  • Combines multiple retrieval methods in a single query runtime
Watch out for
  • No publicly stated pricing for managed or supported versions
  • Requires self-deployment and infrastructure management by users

Omnigraph at a glance

Type
Open-source project
Pricing
Not confirmed. Check the site
Open source
Yes · Source code
Language
English
Last checked
Oct 5, 2026 · Online when checked
Listed
Oct 5, 2026

Omnigraph alternatives

Tools and sites you could use instead, ranked by how closely they match.

All 24 alternatives

REP

github.com/yandex/rep

REP provides an IPython-based machine learning environment for consistent, reproducible data-driven research.

Popular in AI Data & Analytics

Well-known listings from the same topic.

All 302

Omnigraph FAQ

What is Omnigraph?
Omnigraph is an open-source, lakehouse-native graph engine with Git-style workflows for managing graph data and coordinating AI agent fleets.
How do I deploy Omnigraph?
Omnigraph is self-hosted, with installation available via a curl script or Homebrew, and runs on local storage or S3-compatible object stores.
Can multiple AI agents work with Omnigraph simultaneously?
Yes, hundreds of agents can operate on parallel isolated graph branches, with changes reviewed and merged safely via Git-style workflows.
Is Omnigraph open source?
Yes, Omnigraph is open source, with its full source code available in its public GitHub repository.
Is something wrong or out of date?

Tell us and an editor will check it.

Own this website?

Claim the listing and suggest updates yourself. An editor checks every claim first.

Claim this listing