samyama-graph

Overview

Samyama is a Rust-based graph-vector database that can handle massive datasets (up to 74M nodes and 1B edges on a single machine). It implements the OpenCypher query language and provides both Redis protocol support and HTTP endpoints. The project includes a dedicated MCP server that allows AI agents to query knowledge graphs through natural language. Samyama demonstrates impressive performance benchmarks, including querying 1 billion edges for just $2.50. Its MCP integration is particularly notable as it can automatically generate MCP servers from graph schemas, enabling instant AI agent capabilities for any graph dataset.

Try asking AI

After installing, here are 6 things you can ask your AI assistant:

• query_graph

  Execute OpenCypher queries on the graph database

  Try:Find all drugs tested in cancer clinical trialsTry:Who are Alice's friends of friends that work at Google?Try:What is the shortest path between node A and node B?
• run_graph_algorithm

  Execute graph algorithms like PageRank or community detection

  Try:Calculate PageRank for the social network graphTry:Find strongly connected components in the graphTry:Perform triangle counting on the network
• vector_search

  Perform semantic search using vector embeddings

  Try:Find papers similar to this research articleTry:Search for documents related to 'machine learning'Try:Find the most similar nodes based on embeddings
• load_graph_data

  Load domain-specific knowledge graphs into the database

  Try:Load the clinical trials dataset into the graphTry:Import the drug interactions knowledge graphTry:Load the cricket simulation data
• analyze_biomedical_data

  Query biomedical knowledge graphs for research insights

  Try:Find drugs most tested in cancer clinical trialsTry:What are the known side effects of this medication?Try:Find NCI-funded clinical trials for specific drugs
• natural_language_query

  Convert natural language questions to Cypher queries

  Try:Who are Alice's friends of friends that work at Google?Try:Show me the most tested interventions in diabetes trialsTry:Find all articles related to aspirin clinical trials
• create_vector_index

  Create vector indexes for semantic search capabilities

  Try:Create a vector index on Paper nodes with 384 dimensionsTry:Set up cosine similarity index for document embeddingsTry:Build a vector search index for node properties
• execute_algorithm

  Run specific graph algorithms on the database

  Try:Run PageRank algorithm on the social graphTry:Calculate local clustering coefficients for all nodesTry:Perform breadth-first search from a starting node
• analyze_fraud_patterns

  Detect fraud patterns in banking and financial networks

  Try:Find potential money laundering patterns in transactionsTry:Identify clusters of suspicious accountsTry:Trace the flow of high-risk transactions
• analyze_supply_chain

  Analyze supply chain networks for disruptions and optimization

  Try:Identify potential bottlenecks in the supply chainTry:Find alternative routes for disrupted shipping lanesTry:Analyze port congestion impacts on delivery times
• analyze_clinical_trials

  Query clinical trial data for research insights

  Try:Find clinical trials testing specific interventionsTry:Match patients to eligible clinical trialsTry:Analyze adverse events reported in trials
• analyze_manufacturing

  Analyze manufacturing networks for digital twins and failure prediction

  Try:Simulate failure cascades in production linesTry:Identify critical components in manufacturing processesTry:Optimize maintenance schedules based on component dependencies

Note: Tools were inferred from the README's documentation of capabilities and examples. While there's no explicit 'Tools' section, the README demonstrates various functions through examples and feature descriptions. The tools represent the main f

Comparable tools

Installation

# Install Samyama
git clone https://github.com/samyama-ai/samyama-graph && cd samyama-graph
cargo build --release
./target/release/samyama  # Starts RESP on :6379 and HTTP on :8080

# Install the MCP server
pip install samyama[mcp]

# Start an MCP server with demo data
samyama-mcp-serve --demo cricket

{
  "mcpServers": {
    "samyama": {
      "command": "python",
      "args": ["-m", "samyama.mcp"],
      "env": {}
    }
  }
}

FAQ

What query languages does Samyama support?

Samyama supports OpenCypher queries (~90% compatibility), along with a vector search API using Cypher extensions. It also accepts natural language queries that are translated to Cypher.

Can Samyama handle multi-domain knowledge graphs?

Yes, Samyama has loaded and queried multiple interconnected knowledge graphs including PubMed, ClinicalTrials.gov, Reactome, and DrugBank simultaneously on a single machine.

How does the MCP server integration work?

Samyama provides a 'samyama-mcp-serve' command that generates MCP server tools from your graph schema, enabling AI agents to query the database without requiring custom tool definitions.

Compare samyama-graph with