This project demonstrates how to configure multiple ChatModel beans in a Spring application using the new "Programmating Bean Registration" feature introduced in Spring Framework 7.0.

The code included in this project is adapted from the spring-ai-autoconfigure-model-ollama module from the Spring AI project, which provides auto-configuration for Ollama chat models.

The goal of this project is to experiment with the new features and lay the foundation for implementing spring-projects/spring-ai#3518.

• Java 24
• Ollama installed with the qwen3:4b, qwen3:8b, and gemma3:4b models available.

./gradlew bootRun

You can test the application by sending HTTP GET requests to the following endpoints:

• http://localhost:8080/chat - Uses the default chat model Qwen 3 4B (generic chat tasks)
• http://localhost:8080/chat/agent - Uses the Qwen 3 8B model (specific for ficticious agentic tasks)
• http://localhost:8080/chat/rag - Uses the Gemma 3 4B model (specific for ficticious RAG tasks)