deeplethe/utopia

World's first open-source enterprise world model.

What it solves

Utopia addresses the limitations of standard vector stores and knowledge graphs by providing a "bitemporal" knowledge system. Instead of just storing the current state of information, it tracks both when a fact was true in the real world and when the system learned about it. This allows enterprises to maintain a complete audit trail of changing understanding, resolve conflicts in data, and make trustworthy decisions based on a historical record of evidence.

How it works

Built with Rust and Postgres (using pgvector), Utopia transforms unstructured documents (PDFs, Word, etc.) and structured data into entities and facts based on a defined ontology. It uses a bitemporal knowledge graph where corrections do not overwrite old data but close previous versions and link to new ones. The system integrates LLMs for extraction and chat, uses Tantivy for full-text search, and employs "Ontology2SQL" to allow agents to query mounted databases by mapping tables to the ontology. It also includes a human-in-the-loop review queue for low-confidence extractions and entity merges.

Who it’s for

It is designed for enterprise teams—such as legal, procurement, HR, finance, and operations—who need to manage complex records that evolve over time (e.g., amended contracts or changing project leadership) and require a verifiable, air-gapped knowledge foundation for their AI agents.

Highlights

  • Bitemporal Knowledge Graph: Tracks two timelines (world time and system time) to preserve the history of how knowledge evolved.
  • Ontology-Driven: Uses predefined or custom ontology packs to structure data and derive new facts via forward chaining.
  • Agent Integration: Provides an MCP (Model Context Protocol) server, allowing agents like Claude Desktop or Cursor to interact with the knowledge base.
  • Hybrid Search: Combines full-text search (Tantivy) and vector search (pgvector) using Reciprocal Rank Fusion (RRF).
  • Decision Ledger: Maintains an append-only record of every change, merge, or confirmation for compliance and auditing.

Related

  • Project
    a-agmon/rs-graph-llm

    graph-flow is a type-safe, LangGraph-inspired framework for building stateful, resumable AI agent workflows in Rust. It provides graph-based task orchestration with conditional routing, human-in-the-loop pauses, session persistence (in-memory or PostgreSQL), and tight integration with the Rig LLM ecosystem.

  • Project
    siddsachar/row-bot

    Row‑Bot is a self‑hosted desktop AI assistant that lets you converse with LLMs to create files, run code, manage designs, and interact with services—all while keeping data locally. It offers memory via a personal knowledge graph, goal‑driven agents, approval‑gated actions, multi‑device remote access, and integrations (GitHub, Notion, Slack, etc.). Runs on Windows/macOS/Linux or Docker, using Ollama or any OpenAI‑compatible provider, and stores everything securely on your machine.

  • Dispatch
    Upwork OpenAI Integration and AI-First Strategy

    Upwork has transitioned to an OpenAI-centric ecosystem, integrating GPT-4o and GPT-3.5 into its marketplace features, internal fraud detection, and corporate productivity via ChatGPT Enterprise.

  • Project
    FalkorDB/GraphRAG-SDK

    A modular framework for building GraphRAG applications on FalkorDB that improves factual accuracy and reduces LLM hallucinations by using knowledge graphs for traceable, multi-hop evidence retrieval.