Use local Ollama instead of OpenAI

README.md

@@ -1,20 +1,38 @@
 # Minimal Chat application using Spring AI
 
-This project is a Spring Boot application that integrates a Chat UI powered by OpenAI using Vaadin. It provides a responsive interface for interacting with OpenAI's conversational models.
+This project is a Java application that integrates a Chat UI powered by Ollama using Vaadin UI. It provides a responsive interface for interacting with Ollama's conversational models.
 
 <img width="504" alt="vaadin-spring-ai-open-ai-chat" src="https://github.com/samie/spring-ai-chat/assets/991105/d7216455-f28e-4d34-8cd0-f7f1074f6087">
 
 ## Features
 - Full-stack application with Vaadin frontend and Spring Boot backend.
-- Real-time chatting capability using OpenAI.
+- Real-time chatting capability using local Ollama in Docker container.
 - Dynamic message streaming and display of Markdown.
 
 ## Prerequisites
+- Docker container for running ollama models
 - JDK 17 or later
 - Maven 3.6 or later
 
 ## Running the Application
 
+Install and run ollama in localhost:11434
 ```
-OPENAI_API_KEY=<your api key here> mvn spring-boot:run
+docker run -d -v ollama:/root/.ollama -p 11434:11434  --name ollama ollama/ollama
+```
+
+Install ollama 'mistral' model (used by the web app by default)
+```
+docker exec ollama ollama pull mistral
+```
+
+Optional: check that everything works using curl:
+```
+curl http://localhost:11434/api/chat -d '{"model": "mistral", "messages": [{"role": "user", "content": "is black darker than white?"}], "stream":false}'
+```
+
+
+Run the web application in [localhost:8080](http://localhost:8080/):
+```
+mvn spring-boot:run
 ```

pom.xml

@@ -54,7 +54,7 @@
 		</dependency>
 		<dependency>
 			<groupId>org.springframework.ai</groupId>
-			<artifactId>spring-ai-openai-spring-boot-starter</artifactId>
+			<artifactId>spring-ai-ollama-spring-boot-starter</artifactId>
 			<version>0.8.1</version>
 		</dependency>
 		<dependency>