Sunday, 22 June 2025

LangChain4j: Powerful AI & LLM Features for Java Developers

 

LangChain4j brings the full power of modern AI to Java applications with a rich set of features.

 

1. Integrations

 

15+ LLM Providers (OpenAI, Gemini, Claude, Mistral, and more)

·      Amazon Bedrock (Converse API)

·      Amazon Bedrock (Invoke API)

·      Anthropic

·      Azure OpenAI

·      ChatGLM

·      DashScope

·      GitHub Models

·      Google AI Gemini

·      Google Vertex AI Gemini

·      Google Vertex AI PaLM 2

·      Hugging Face

·      Jlama

·      LocalAI

·      Mistral AI

·      Ollama

·      OpenAI

·      Qianfan

·      Cloudflare Workers AI

·      Zhipu AI

 

You can get the latest list here (https://docs.langchain4j.dev/integrations/language-models/)

 

 

20+ Vector Stores (Pinecone, Chroma, Redis, Weaviate, etc.)

Following is the list of vectors embedding stores supported:

·      In-memory

·      AlloyDB for Postgres

·      Astra DB

·      Azure AI Search

·      Azure CosmosDB Mongo vCore

·      Azure CosmosDB NoSQL

·      Cassandra

·      Chroma

·      ClickHouse

·      Cloud SQL for Postgres

·      Coherence

·      Couchbase

·      DuckDB

·      Elasticsearch

·      Infinispan

·      Milvus

·      MongoDB Atlas

·      Neo4j

·      OpenSearch

·      Oracle

·      PGVector

·      Pinecone

·      Qdrant

·      Redis

·      Tablestore

·      Vearch

·      Vespa

·      Weaviate

 

Refer this link (https://docs.langchain4j.dev/integrations/embedding-stores/) to get latest stores supported by Langchain4j.

 

 

15+ Embedding Models (OpenAI, Hugging Face, Cohere, etc.)

·      In-process (ONNX)

·      Amazon Bedrock

·      Azure OpenAI

·      Cohere

·      DashScope

·      GitHub models

·      Google Vertex AI

·      Hugging Face

·      Jina

·      Jlama

·      LocalAI

·      Mistral AI

·      Nomic

·      Ollama

·      OpenAI

·      OpenAI Official SDK

·      OVHcloud AI Endpoints

·      Qianfan

·      Voyage AI

·      Cloudflare Workers AI

·      Zhipu AI

·      Xinference

 

You can get latest adoption here (https://docs.langchain4j.dev/category/embedding-models/).

 

 

5+ Image Generation Models (DALL·E, Stable Diffusion, etc.)

·      Azure OpenAI Dall.E

·      OpenAI Dall.E

·      Google Imagen

·      OpenAI Official SDK

·      Cloudflare Workers AI

·      ZhiPu AI

·      Xinference

 

You can get latest adoption here (https://docs.langchain4j.dev/category/image-models/)

 

 

5+ Scoring and reranking Models (for re-ranking retrieval results)

·      In-process (ONNX)

·      Cohere

·      Jina

·      Google Cloud Vertex AI Ranking API

·      Voyage AI

·      Xinference

 

You can get latest adoption here (https://docs.langchain4j.dev/category/scoring-reranking-models/)

 

2. Document Processing & RAG

·      Import from multiple sources (Files, URLs, GitHub, S3, Azure Blob)

·      Support for various formats (PDF, DOCX, PPT, XLS, TXT, etc.)

·      Smart text splitting (multiple segmentation strategies)

·      Embedding generation & vector storage

·      Advanced retrieval (query expansion, compression, hybrid search)

·      Re-ranking & Reciprocal Rank Fusion

 

 

3. AI Services & Tools

·      High-level LLM API (simplifies complex AI workflows)

·      Prompt templating (dynamic prompt generation)

·      Function calling (Tools) (execute custom logic via LLM)

·      Dynamic Tools (run dynamically generated code from LLMs)

·      Text classification

 

4. Chat & Memory Management

·      Persistent & in-memory chat memory

·      Message & token windowing (control context size)

·      Streaming responses (real-time LLM output handling)

 

5. Utilities

·      Output parsing (to Java objects & custom POJOs)

·      Token counting & estimation

·      Kotlin coroutine support (async, non-blocking operations)

 

6. Multimodal Support

·      Text & image inputs (for vision-capable models)

 

LangChain4j makes it easy to build AI-powered Java apps from chatbots to RAG systems without leaving the JVM ecosystem.

 

Previous                                                    Next                                                    Home

No comments:

Post a Comment