{"id":6806,"date":"2025-09-19T01:34:18","date_gmt":"2025-09-19T01:34:18","guid":{"rendered":"https:\/\/techtrendfeed.com\/?p=6806"},"modified":"2025-09-19T01:34:19","modified_gmt":"2025-09-19T01:34:19","slug":"adk-for-java-opening-as-much-as-third-party-language-fashions-by-way-of-langchain4j-integration","status":"publish","type":"post","link":"https:\/\/techtrendfeed.com\/?p=6806","title":{"rendered":"ADK for Java opening as much as third-party language fashions by way of LangChain4j integration"},"content":{"rendered":"


\n<\/p>\n

\n

\"adk-langchain4j-banner\"\/ <\/p>\n

\n

The current 0.2.0 launch<\/a> of Google\u2019s Agent Growth Equipment<\/b> (ADK) for Java<\/b> provides an integration with the LangChain4j<\/a> LLM framework. This integration offers builders with a big selection of Massive Language Fashions<\/b> (LLMs) supported by LangChain4j, for constructing AI brokers.<\/p>\n

Along with ADK\u2019s built-in Google Gemini and Anthropic Claude integrations, builders can now use LangChain4j to entry different fashions from third-party suppliers (like OpenAI, Anthropic, GitHub, Mistral…) or native open-weight fashions, e.g. by way of Ollama<\/a> or Docker Mannequin Runner<\/a>.<\/p>\n<\/div>\n

\n

LangChain4j integration for a wide selection of fashions<\/h2>\n

The LangChain4j LLM framework helps all kinds of fashions. You’ll be able to test the checklist of supported fashions<\/a> within the LangChain4j documentation. Let\u2019s take a look at a pair concrete examples, utilizing Gemma with Docker Mannequin Runner, and Ollama with Qwen.<\/p>\n

When declaring your ADK agent with the LlmAgent<\/code> builder, you specify the LLM by way of the mannequin()<\/code> builder methodology. You normally go a string representing the identify of the mannequin, like “gemini-2.5-flash<\/code>“.<\/p>\n

It\u2019s additionally potential to make use of an occasion of a category extending the BaseLlm<\/code> summary class. That is precisely what the mixing with LangChain4j does, to create a bridge between each frameworks. It’s important to use a brand new LangChain4j<\/code> class that extends this BaseLlm<\/code> class.<\/p>\n<\/div>\n

\n

Operating Gemma 3 with Docker Mannequin Runner<\/h3>\n

After having put in and enabled Docker Mannequin Runner<\/a> in your machine, you may pull the Gemma 3 mannequin simply by way of this command:<\/p>\n<\/div>\n

\n
docker mannequin pull ai\/gemma3<\/code><\/pre>\n
\n        Shell\n    <\/p>\n<\/div>\n
\n
As Docker Mannequin Runner fashions exposes an OpenAI suitable API floor, you need to use the LangChain4j module for OpenAI suitable fashions, by specifying the next dependencies in your Maven pom.xml<\/code>:<\/p>\n<\/div>\n
\n

\n    com.google.adk<\/groupid>
\n    google-adk-contrib-langchain4j<\/artifactid>
\n    0.2.0<\/version>
\n<\/dependency>
\n
\n    dev.langchain4j<\/groupid>
\n    langchain4j-open-ai<\/artifactid>
\n    1.4.0<\/version>
\n<\/dependency><\/code><\/pre>\n
\n        XML\n    <\/p>\n<\/div>\n
\n
Then, create a LangChain4j chat mannequin, specifying the mannequin you need to use, and the native URL and port:<\/p>\n<\/div>\n
\n
OpenAiChatModel dmrChatModel = OpenAiChatModel.builder()
\n    .baseUrl(\"http:\/\/localhost:12434\/engines\/llama.cpp\/v1\")
\n    .modelName(\"ai\/gemma3n\")
\n    .construct();<\/code><\/pre>\n
\n        Java\n    <\/p>\n<\/div>\n
\n
Now, configure a chess coach agent utilizing that mannequin:<\/p>\n<\/div>\n
\n
LlmAgent chessCoachAgent = LlmAgent.builder()
\n    .identify(\"chess-coach\")
\n    .description(\"Chess coach agent\")
\n    .mannequin(new LangChain4j(dmrChatModel))
\n    .instruction(\"\"\"
\n        You're a educated chess coach
\n        who helps chess gamers practice and sharpen their chess abilities.
\n        \"\"\")
\n    .construct();<\/code><\/pre>\n
\n        Java\n    <\/p>\n<\/div>\n
\n
Discover how the bridge between the 2 frameworks is completed by way of the mannequin(new LangChain4j(dmrChatModel))<\/code> instruction. And right here you go, your AI agent is powered by an area mannequin!<\/p>\n<\/div>\n
\n
Operating Qwen 3 with Ollama<\/h3>\n
If as a substitute you need to construct a pleasant science instructor agent with the Qwen 3<\/a> mannequin working regionally in your machine by way of Ollama<\/a>, first, outline our dependencies inside a Maven pom.xml construct file:<\/p>\n<\/div>\n
\n

\n
\n    com.google.adk<\/groupid>
\n    google-adk<\/artifactid>
\n    0.2.0<\/version>
\n<\/dependency>
\n
\n
\n    com.google.adk<\/groupid>
\n    google-adk-contrib-langchain4j<\/artifactid>
\n    0.2.0<\/version>
\n<\/dependency>
\n
\n
\n    dev.langchain4j<\/groupid>
\n    langchain4j-ollama<\/artifactid>
\n    1.4.0<\/version>
\n<\/dependency><\/code><\/pre>\n
\n        XML\n    <\/p>\n<\/div>\n
\n
Let\u2019s assume you\u2019ve already put in Ollama<\/a> in your machine, you pulled the Qwen 3 mannequin, and received it working on port 11434<\/code>. With LangChain4j, in Java, you instantiate the Ollama mannequin supplier as follows:<\/p>\n<\/div>\n
\n
OllamaChatModel ollamaChatModel = OllamaChatModel.builder()
\n    .modelName(\"qwen3:1.7b\")
\n    .baseUrl(\"http:\/\/127.0.0.1:11434\")
\n    .construct();<\/code><\/pre>\n
\n        Java\n    <\/p>\n<\/div>\n
\n
Now let\u2019s wire this mannequin right into a easy science instructor agent:<\/p>\n<\/div>\n
\n
LlmAgent scienceTeacherAgent = LlmAgent.builder()
\n    .identify(\"science-app\")
\n    .description(\"Science instructor agent\")
\n    .mannequin(new LangChain4j(ollamaChatModel))
\n    .instruction(\"\"\"
\n        You're a useful science instructor
\n        who explains science ideas to youngsters and youngsters.
\n        \"\"\")
\n    .construct();<\/code><\/pre>\n
\n        Java\n    <\/p>\n<\/div>\n
\n
If the mannequin helps perform calling, you may give your agent entry to instruments as properly. For instance, give it entry to MCP servers, or your native code pushed features. You’ll be able to discover the assorted instruments at your disposal on this article diving into ADK instruments<\/a>, or by wanting on the ADK documentation<\/a>.<\/p>\n<\/div>\n
\n
New Options on this Launch<\/h2>\n
Past the headline LangChain4j integration, model 0.2.0 brings a number of different highly effective enhancements to the agent growth workflow:<\/p>\n