Experience NeuronDB ecosystem capabilities firsthand. Try interactive terminal demos for vector search, ML inference, agent runtime, and MCP protocol.
$ Try the demos below to see NeuronDB in action...NeuronDB Demo
Vector search, ML inference, and RAG pipeline demonstrations
This demo shows NeuronDB capabilities including vector search with HNSW indexing, machine learning inference with ONNX models, embedding generation, hybrid search combining semantic and full-text search, RAG pipeline operations, and GPU acceleration. SQL commands execute to demonstrate vector similarity search, ML model training and prediction, document processing, and more in PostgreSQL.
NeuronAgent Demo
REST API and agent runtime demonstrations
This demo shows NeuronAgent REST API endpoints for building AI agent systems. You will see how to create agents with specific profiles and tools, establish sessions for conversation management, send messages that trigger agent actions using NeuronDB vector search and embeddings, and retrieve responses. The demo shows the agent lifecycle including health checks and API authentication.
NeuronMCP Demo
MCP protocol and Claude Desktop integration demonstrations
This demo shows NeuronMCP Model Context Protocol server implementation. You will see JSON-RPC 2.0 protocol messages over stdio communication, showing how MCP-compatible clients like Claude Desktop can interact with NeuronDB. The demo includes tool discovery, vector search operations, embedding generation, resource management, and how the MCP server provides access to NeuronDB vector search, ML algorithms, and RAG capabilities through a standardized protocol interface.