Spring AI Playground is an innovative self-hosted web UI designed for Java developers aiming to simplify AI experimentation and testing. This platform provides an intuitive interface for interacting with large language models (LLMs), managing vector databases, engaging in prompt engineering, and integrating with Model Context Protocol (MCP).

With a focus on accessibility, Spring AI Playground is built on the foundational elements of Spring AI to enable developers to quickly prototype Spring AI-based applications. The platform supports leading AI model providers and facilitates comprehensive testing of retrieval-augmented generation (RAG) workflows alongside various MCP integrations. By utilizing this playground, developers gain enhanced contextual awareness and can leverage external tool capabilities in their applications.

Key Features

• AI Model Support: Integrate with major AI model providers including OpenAI, Anthropic, Microsoft, and more.
• MCP Playground: Manage connections to external tools, test and configure MCP server connections with ease, and explore available tools through the intuitive MCP inspector.
• Vector Database Integration: Effortlessly switch between various vector database providers, perform vector similarity searches, and visualize results with Spring AI's VectorStore API.
• RAG Implementation: Augment AI responses by retrieving information from user-uploaded documents, enhancing the relevance and contextuality of generated outputs.
• Progressive Web App (PWA): Experience app-like functionality with the ability to install the playground as a standalone application on devices.

Quick Start

Prerequisites

• Java 21 or later installed (required for building the project).
• Ollama running on your machine (refer to AI Models).
• Docker installed and running on your machine. (only if you choose to run the application using Docker)

Getting Started

First, clone the Spring AI Playground project from GitHub:

git clone https://github.com/JM-Lab/spring-ai-playground.git
cd spring-ai-playground

Running the Application

Running with Docker (Recommended)

1. Build the Docker Image:

   ./mvnw spring-boot:build-image -Pproduction -DskipTests=true \
   -Dspring-boot.build-image.imageName=jmlab/spring-ai-playground:latest
   

2. Run the Docker Container:

   docker run -d -p 8282:8282 --name spring-ai-playground \
   -e SPRING_AI_OLLAMA_BASE_URL=http://host.docker.internal:11434 \
   -v spring-ai-playground:/home \
   --restart unless-stopped \
   jmlab/spring-ai-playground:latest
   

3. Access the Application: Open http://localhost:8282 in your browser.

Running Locally (Optional)

1. Build and Run the Application:

   ./mvnw clean install -Pproduction -DskipTests=true
   ./mvnw spring-boot:run
   

2. Access the Application: Open http://localhost:8282 in your browser.

In addition, developers can configure their environment to test different AI models and manage integrations seamlessly, helping them innovate without barriers.

For further exploration, check the upcoming features that promise exciting developments, including enhanced agent capabilities and multimodal support for richer AI interactions.