spring-ai

Accelerator to kick-start development of applications using Spring AI, Ollama and DeepSeek/Mistral. We develop a simple RAG (Retrieval-Augmented Generation) system using DeepSeek/Mistral and PostgreSQL Vector Database to chat with PDFs and answer complex questions about our local documents.

We use Spring Boot Data JDBC to connect to the Postgre Vector Database which is launched locally with Docker Compose. Docker Compose requires the Docker engine to be already installed (our codespace has it).

Prerequisites

The Codespaces for this repo pre-installs Ollama. However, after you have launched CodeSpaces, pull the Deepseek model deepseek-r1:1.5b which is 1.1 GB in size using the command below.

ollama pull deepseek-r1:1.5b

For the RAG (Retrieval-Augmented Generation) implementation, pull the latest Mistral model which is 4.1 GB in size using the command below.

ollama pull mistral:latest

You should be able to list the models and the Docker image for PostgreSQL 16.4 as shown below:

@kulkeez ➜ /workspaces/spring-ai (master) $ ollama list
NAME                ID              SIZE      MODIFIED    
mistral:latest      f974a74358d6    4.1 GB    2 hours ago    
deepseek-r1:1.5b    a42b25d8c10a    1.1 GB    7 hours ago    
@kulkeez ➜ /workspaces/spring-ai (master) $ docker images
REPOSITORY          TAG       IMAGE ID       CREATED        SIZE
pgvector/pgvector   pg16      c0fb89364d78   3 months ago   435MB
@kulkeez ➜ /workspaces/spring-ai (master) $ 

---

Want to run it?

This Spring AI application can be launched from command-line by the developer using the command: mvn spring-boot:run On launching the application the first time, the Postgre Vector database Docker image is pulled and started.

---

Query this AI application

Launch any web browser and invoke this application with a prompt text. Example: http://localhost:8080/twister?message=Tell me a tongue twister

You can use HTTPie to check if the application is ready.

@kulkeez ➜ /workspaces/spring-ai (master) $ http :8080/
HTTP/1.1 200 
Connection: keep-alive
Content-Type: application/json
Date: Sat, 01 Feb 2025 14:01:58 GMT
Keep-Alive: timeout=60
Transfer-Encoding: chunked

{
    "Date": "Sat Feb 01 14:01:58 UTC 2025",
    "Message": "Welcome to Generative AI world using Spring AI!"
}


@kulkeez ➜ /workspaces/spring-ai (master) $ 

---

Debugging

If the application doesn't boot up and throws SQL Exceptions, you can review the PostGres database by logging into the Docker container using the following commands to recreate the Table and index:

@kulkeez ➜ /workspaces/spring-ai (master) $ docker ps 
CONTAINER ID   IMAGE                    COMMAND                  CREATED          STATUS          PORTS                                           NAMES
439dcc7172e6   pgvector/pgvector:pg16   "docker-entrypoint.s…"   43 minutes ago   Up 25 seconds   0.0.0.0:32768->5432/tcp, [::]:32768->5432/tcp   pgvector
@kulkeez ➜ /workspaces/spring-ai (master) $
@kulkeez ➜ /workspaces/spring-ai (master) $ docker exec -it pgvector bash
root@b4324bb6f9f8:/# psql -h localhost -U postgres --password -d aidemo
Password: 
psql (16.4 (Debian 16.4-1.pgdg120+2))
Type "help" for help.

aidemo=# 
aidemo-# 
aidemo-# \d vector_store
 id        | uuid         |           | not null | uuid_generate_v4()
 content   | text         |           |          | 
 metadata  | json         |           |          | 
 embedding | vector(4096) |           |          | 

aidemo-# 
aidemo=# drop table vector_store;
DROP TABLE
aidemo=#
aidemo=# CREATE TABLE IF NOT EXISTS vector_store (
        id uuid DEFAULT uuid_generate_v4() PRIMARY KEY,
        content text,
        metadata json,
        embedding vector(1536)
);
CREATE TABLE
aidemo=#
aidemo=# CREATE INDEX IF NOT EXISTS spring_ai_vector_index ON vector_store USING HNSW (embedding vector_cosine_ops);
CREATE INDEX
aidemo=#

Further Learning

Books

• "Spring AI in Action" by Craig Walls (Manning)

Videos

• RAG implementation in Java using Spring AI
• Java RAG made easy with SpringAI and Elasticsearch
• Use RAG developed in Python to chat with PDFs using Deepseek, Langchain and Streamlit
• How to Use DeepSeek's R1 Model in Java & Spring | Local & API Implementation

Spring AI Documentation

• Spring AI Project
• Spring AI Reference Documentation
• Spring AI API Documentation
• Spring AI Playground - A web UI designed to make it easy for Java developers to experiment with and integrate AI models.

Articles

• Spring AI with Vector Databases - Integrating Spring AI with Vector Databases - A Guide Using PGVector
• Create a RAG (Retrieval Augmented Generation) Application with Redis and Spring AI

Tools

• HTTPie - HTTPie (pronounced aitch-tee-tee-pie) is a command-line HTTP client.