: A simple and popular Java library (wrapper) for the Ollama server. It supports:
: Your Java code sends prompts to the Ollama server. If a requested model isn't present, Ollama can be configured to pull it automatically from its library.
In your application.properties or application.yml , configure the model: properties
was a ghost. He lived in the "Ollamac" project—a code-named initiative meant to bridge the gap between Large Language Models and enterprise Java environments. It was supposed to be a tool for efficiency, but for Elias, it had become a cathedral. ollamac java work
: Register standard Java methods as "tools" that the model can choose to call. The library handles the automatic conversion of Java methods into JSON schemas for the model.
Build customer service bots for restricted networks.
Ollama runs a local API (usually on port 11434). Since Java doesn't have a native "Ollama client" built into the standard library, you have two main ways to make them work together: : A simple and popular Java library (wrapper)
Tool calling enables the model to request the execution of a specific function. For example, in a customer service chatbot, the model might identify a user's intent to check an order status and respond by asking your code to call a getOrderStatus(orderId) API. The model returns a structured JSON object specifying the tool to use and its arguments. Spring AI provides robust abstractions for simplifying tool calling.
: Creating desktop or web-based chatbots that run entirely offline.
"model": "qwen2.5:7b", "prompt": "%s", "stream": false In your application
: Your AI-powered features will work even without a constant internet connection. Core Integration Strategies
import dev.langchain4j.model.ollama.OllamaChatModel; public class LocalAiAssistant public static void main(String[] args) // Configure the model to match what you downloaded in Ollamac OllamaChatModel model = OllamaChatModel.builder() .baseUrl("http://localhost:11434") .modelName("llama3") .temperature(0.7) .build(); System.out.println("Thinking..."); String response = model.generate("Explain the benefit of running local LLMs in one sentence."); System.out.println("\nAI Response:"); System.out.println(response); Use code with caution. Method 2: The Lightweight Approach (Spring AI)
Using spring-ai-ollama-spring-boot-starter you can autowire an EmbeddingModel bean. This provides a simple, high-level abstraction for generating embeddings, which you can then use to search for relevant chunks of text in your vector store.