[![PyPI version](https://img.shields.io/pypi/v/trustgraph.svg)](https://pypi.org/project/trustgraph/) ![E2E Tests](https://github.com/trustgraph-ai/trustgraph/actions/workflows/release.yaml/badge.svg) [![Discord](https://img.shields.io/discord/1251652173201149994 )](https://discord.gg/sQMwkRz5GX) [![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/trustgraph-ai/trustgraph) [**Docs**](https://docs.trustgraph.ai) | [**YouTube**](https://www.youtube.com/@TrustGraphAI?sub_confirmation=1) | [**Configuration Builder**](https://config-ui.demo.trustgraph.ai/) | [**Discord**](https://discord.gg/sQMwkRz5GX) | [**Blog**](https://blog.trustgraph.ai/subscribe)
# Intelligent Context for LLMs TrustGraph provides an event-driven data infrastructure that transforms data into LLM-ready intelligent context through automated knowledge graph construction, vector embeddings mapping, and structuring data using custom ontologies and schemas — all deployable privately, on-prem, or in cloud. Deploy and manage open LLMs with the platform, ensuring complete data sovereignty while enabling intelligent agents that generate real, actionable insights.
Table of Contents
- [**Key Features**](#key-features)
- [**Why TrustGraph?**](#why-trustgraph)
- [**Agentic MCP Demo**](#agentic-mcp-demo)
- [**Getting Started**](#getting-started)
- [**Configuration Builder**](#configuration-builder)
- [**Knowledge Cores**](#knowledge-cores)
- [**Integrations**](#integrations)
- [**Observability & Telemetry**](#observability--telemetry)
- [**Contributing**](#contributing)
- [**License**](#license)
- [**Support & Community**](#support--community)
## Key Features TrustGraph is not just another AI framework but a complete, production-ready platform that bridges the gap between raw data and intelligent, adaptable agent deployments. - **Data Ingestion** *PDF parsing, OCR, configurable chunking, custom metadata, and bring your own schemas for large scale data ingestion*. - **Data Transformation** *Transform unstructured data into flat knowledge graphs or bring your own ontology for rich graphs*. - **Context Retrieval** *Deterministic knowledge graph queries, no LLMs in retrieval path*. - **Event-Driven** *All services communicate through Apache Pulsar topics rather than direct RPC.* - **Multi-Backend Storage** *Data storage with Apache Cassandra, Neo4j, Qdrant, Milvus, Memgraph, FalkorDB, and Pinecone*. - **Data Sovereignty** *Deploy the entire stack—data pipelines, knowledge graphs, vector stores, and LLMs—on-premises, in your VPC, or across hybrid environments*. - **Private LLM Inferencing** *Support for all major LLM APIs or deploy and manage open models directly integrated with the platform*. - **Three-Dimensional Multi-Tenancy** *3 layers of isolation with flows for processing logic, collections for data access, and tool groups for multi-agent.* - **Container-First** *Containerized deployments with Docker or Kubernetes. Built for enterprise scale with monitoring, observability, and management*. - **MCP Integration** *Native support for MCP enables standardized agent communication with third-party tools and services while maintaining data sovereignty*. - **Observable by Design** *3D visualization of knowledge graphs. Prometheus metrics and Grafana dashboards track latency, throughput, costs, and errors*. ## Why TrustGraph? [![Why TrustGraph?](https://img.youtube.com/vi/Norboj8YP2M/maxresdefault.jpg)](https://www.youtube.com/watch?v=Norboj8YP2M) ## Agentic MCP Demo [![Agentic MCP Demo](https://img.youtube.com/vi/mUCL1b1lmbA/maxresdefault.jpg)](https://www.youtube.com/watch?v=mUCL1b1lmbA) ## Getting Started - [**Quickstart Guide**](https://docs.trustgraph.ai/getting-started/) - [**Configuration Builder**](#configuration-builder) - [**Workbench**](#workbench) - [**Developer APIs and CLI**](https://docs.trustgraph.ai/reference/) - [**Deployment Guide**](https://docs.trustgraph.ai/deployment/) ### Watch TrustGraph 101 [![TrustGraph 101](https://img.youtube.com/vi/rWYl_yhKCng/maxresdefault.jpg)](https://www.youtube.com/watch?v=rWYl_yhKCng) ## Configuration Builder The [**Configuration Builder**](https://config-ui.demo.trustgraph.ai/) assembles all of the selected components and builds them into a deployable package. It has 4 sections: - **Version**: Select the version of TrustGraph you'd like to deploy - **Component Selection**: Choose from the available deployment platforms, LLMs, graph store, VectorDB, chunking algorithm, chunking parameters, and LLM parameters - **Customization**: Enable OCR pipelines and custom embeddings models - **Finish Deployment**: Download the launch `YAML` files with deployment instructions ## Workbench The **Workbench** is a UI that provides tools for interacting with all major features of the platform. The **Workbench** is enabled by default in the **Configuration Builder** and is available at port `8888` on deployment. The **Workbench** has the following capabilities: - **Agentic, GraphRAG and LLM Chat**: Chat interface for agentic flows, GraphRAG queries, or directly interfacing with a LLM - **Semantic Discovery**: Analyze semantic relationships with vector search, knowledge graph relationships, and 3D graph visualization - **Data Management**: Load data into the **Librarian** for processing, create and upload **Knowledge Packages** - **Flow Management**: Create and delete processing flow patterns - **Prompt Management**: Edit all LLM prompts used in the platform during runtime - **Agent Tools**: Define tools used by the Agent Flow including MCP tools - **MCP Tools**: Connect to MCP servers ## Knowledge Cores A challenge facing RAG architectures is the ability to quickly reuse and remove datasets from pipelines. **TrustGraph** stores the results of the data ingestion process in reusable Knowledge Cores. Knowledge cores can be loaded and removed during runtime. Some sample knowledge cores are [here](https://github.com/trustgraph-ai/catalog/tree/master/v3). A Knowledge Core has two components: - Knowledge graph triples - Vector embeddings mapped to the knowledge graph ## Integrations TrustGraph provides maximum flexibility to avoid vendor lock-in.
LLM APIs
- Anthropic
- AWS Bedrock
- AzureAI
- AzureOpenAI
- Cohere
- Google AI Studio
- Google VertexAI
- Mistral
- OpenAI
LLM Orchestration
- LM Studio
- Llamafiles
- Ollama
- TGI
- vLLM
VectorDBs
- Qdrant (default)
- Pinecone
- Milvus
Graph Storage
- Apache Cassandra (default)
- Neo4j
- Memgraph
- FalkorDB
Observability
- Prometheus
- Grafana
Control Plane
- Apache Pulsar
Clouds
- AWS
- Azure
- Google Cloud
- OVHcloud
- Scaleway
## Observability & Telemetry Once the platform is running, access the Grafana dashboard at: ``` http://localhost:3000 ``` Default credentials are: ``` user: admin password: admin ``` The default Grafana dashboard tracks the following:
Telemetry
- LLM Latency
- Error Rate
- Service Request Rates
- Queue Backlogs
- Chunking Histogram
- Error Source by Service
- Rate Limit Events
- CPU usage by Service
- Memory usage by Service
- Models Deployed
- Token Throughput (Tokens/second)
- Cost Throughput (Cost/second)
## Contributing [Developer's Guide](https://docs.trustgraph.ai/community/developer.html) ## License **TrustGraph** is licensed under [Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0). Copyright 2024-2025 TrustGraph Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. ## Support & Community - Bug Reports & Feature Requests: [Discord](https://discord.gg/sQMwkRz5GX) - Discussions & Questions: [Discord](https://discord.gg/sQMwkRz5GX) - Documentation: [Docs](https://docs.trustgraph.ai/)