utopia

World's first open-source enterprise world model.

What is utopia?

World's first open-source enterprise world model.

Overview of utopia

What it is

Utopia is an open-source enterprise world model developed by DeepLethe. It is described as the first open substrate for knowledge engineering that learns passively and governs itself. Unlike traditional knowledge graphs or vector stores that hold present knowledge, Utopia incorporates time awareness and ontology into its base layer, allowing the knowledge system to evolve as new material arrives. It supports conflict detection, reasoning, and decision-making against that ontology. The system deploys offline, enabling companies to build a knowledge foundation, a decision core for agents, and a compliance audit trail on their own hardware.

Key features

  • A complete application with a system console, graph browser, and ontology workbench in one web UI.
  • Knowledge ingest supporting PDF, DOCX, PPTX, XLSX, XLS, ODS, CSV, TSV, Markdown, HTML, plain text, and more; scans and images via MinerU or Volcengine Ark; recordings via transcription; web pages, RSS, GitHub, Jira, Notion, WebDAV, and S3 sync on schedule.
  • Search and chat with full-text search on Tantivy, vectors on pgvector, fused with RRF; answers stream with inline citations; compatible with any OpenAI-compatible endpoint (DeepSeek, Qwen, GLM, Ollama, vLLM) for air-gapped operation.
  • Agent harness and agentic RAG; built-in agent searches documents, walks the graph, and queries databases; read-only tools exposed over MCP.
  • Bitemporal knowledge graph: every fact carries when it held and where it came from; corrections close old versions and link new ones without overwriting.
  • Entity resolution and review in three stages (exact name, embedding similarity, model call); uncertain cases go to a review queue.
  • Conflict detection for three types: new fact vs. old fact, axiom violations (self-loop, asymmetry, transitive cycle, cardinality), and contradictory derivations.

Use cases

Utopia is designed for scenarios where temporal accuracy and provenance matter. Use cases include:

  • Contract chain management: tracking payment terms or parties across amendments over time.
  • Who held what, when: tracking roles, corrections, and restatements in HR, finance, or research records.
  • Agent that reads the base: connecting Claude Desktop, Cursor, or other agents over MCP to query facts with stable identifiers and evidence.
  • Observations from a system: ingesting time-stamped observations from robotics, monitoring, or security tooling.

Getting started

See the repository README for installation and quick-start instructions.

License and pricing

Licensed under Apache-2.0. The project is open source.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/deepletheutopia.svg)](https://yunodb.com/tools/deepletheutopia?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/deepletheutopia?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/deepletheutopia.svg" alt="Featured on YunoDB" height="20"></a>