spring-ai-intro

Gentle introduction to Spring AI.

How to run

./mvnw spring-boot:run

Chat Client API Examples

All chat examples are packaged in chat package, e.g.

• SimpleChatController
• PromptChatController
• RoleController
• MultiModalController

SimpleChatController

SimpleChatController demonstrates a simple ChatClient API that connects to your LLM model.

http localhost:8080/chat/simple

http 'localhost:8080/chat/simple?query=tell-me-a-joke'

PromptChatController

PromptChatController demonstrates simple prompt templating.

http 'localhost:8080/chat/joke'

http 'localhost:8080/chat/joke?topic=ai'

http 'localhost:8080/chat/joke?lang=swedish'

RoleController

RoleController demonstrates use of user and system prompts.

http localhost:8080/chat/fruit

http localhost:8080/chat/veggie

MultiModalController

MultiModalController demonstrates usage of Multimodality API and how it can simultaneously understand and process from various sources, including text, images, audio, video, etc.

This will work on OpenAI, but not on Ollama llama3.2 model as it is a text-based AI.

http localhost:8080/chat/explain

RAG Example

LoadController processes JSON objects and stores their embeddings in a vector store.

http localhost:8080/raffle/load/json

RaffleController leverages the data in the vector store and answers questions about the stored data.

http localhost:8080/raffle/chat

http 'localhost:8080/raffle/chat?question=list all positive feedback'

http 'localhost:8080/raffle/chat?question=list all negative feedback'