About LangChat
LangChat (opens in a new tab) is an enterprise AI knowledge base and large model application solution based on the Java ecosystem, helping enterprises quickly build AI large model applications. Additionally, LangChat integrates an RBAC permissions system, providing enterprises with an out-of-the-box AI large model product solution.
Technology Stack
LangChat (opens in a new tab) leverages the Java ecosystem with a separated front-end and back-end architecture and utilizes the latest technology stack. The back-end is based on SpringBoot3, while the front-end uses Vue3. LangChat not only provides product solutions in the AI domain for enterprises but also serves as a complete Java enterprise application case. This system will guide you through the full development process of SpringBoot3 and Vue3, business modularization, and AI application integration. Whether for enterprise development or personal learning, LangChat offers numerous learning cases. The involved technology stack includes:
Back-end:
- SpringBoot: MVC framework
- Mybatis Plus: Persistence framework
- Sa-Token: Authorization framework
- Hutool: Java utility library
- LiteFlow: Java process engine
- AI LLM...
Front-end:
- Vue3
- TypeScript
- Node
- EChart
- NaiveUI
- vue-flow
LangChain4j
The AI core implementation of LangChat is based on LangChain4j (opens in a new tab). LangChain4j is LangChain's solution in the Java ecosystem. Similar to LangChain's design philosophy, LangChain4j provides various out-of-the-box abstractions for LLMs, simplifying their integration with Java applications.
Various function calls, advanced RAG, and web search functionalities of LangChat are based on LangChain4j. For more features, please refer to the LangChain official documentation: https://docs.langchain4j.dev/ (opens in a new tab)
Supported AI Models for Integration
Provider | Native Image | Sync Completion | Streaming Completion | Embedding | Image Generation | Scoring | Function Calling |
---|---|---|---|---|---|---|---|
OpenAI | ✅ | ✅ | ✅ | ✅ | ✅ | ||
Azure OpenAI | ✅ | ✅ | ✅ | ✅ | ✅ | ||
Hugging Face | ✅ | ✅ | |||||
Amazon Bedrock | ✅ | ✅ | ✅ | ||||
Google Vertex AI Gemini | ✅ | ✅ | ✅ | ✅ | |||
Google Vertex AI | ✅ | ✅ | ✅ | ✅ | |||
Mistral AI | ✅ | ✅ | ✅ | ✅ | |||
DashScope | ✅ | ✅ | ✅ | ✅ | |||
LocalAI | ✅ | ✅ | ✅ | ✅ | |||
Ollama | ✅ | ✅ | ✅ | ||||
Cohere | ✅ | ||||||
Qianfan | ✅ | ✅ | ✅ | ✅ | |||
ChatGLM | ✅ | ||||||
Nomic | ✅ | ||||||
Anthropic | ✅ | ✅ | |||||
Zhipu AI | ✅ | ✅ | ✅ | ✅ |
Supported Vector Databases for Integration
Provider | Storing Metadata | Filtering by Metadata | Removing Embeddings |
---|---|---|---|
In-memory | ✅ | ✅ | ✅ |
Astra DB | ✅ | ||
Azure AI Search | ✅ | ||
Azure CosmosDB Mongo vCore | ✅ | ||
Cassandra | ✅ | ||
Chroma | ✅ | ||
Elasticsearch | ✅ | ✅ | ✅ |
Infinispan | ✅ | ||
Milvus | ✅ | ✅ | ✅ |
MongoDB Atlas | ✅ | ||
Neo4j | |||
OpenSearch | ✅ | ||
PGVector | ✅ | ✅ | ✅ |
Pinecone | |||
Qdrant | ✅ | ||
Redis | ✅ | ||
Vearch | ✅ | ||
Vespa | |||
Weaviate | ✅ | ✅ |