Repolode
Submit
davidmigloz

davidmigloz/langchain_dart

A Dart and Flutter port of LangChain for building LLM-powered apps with pluggable model providers, RAG, and agents.

View on GitHubWebsite
Stars
689
Forks
156
Open issues
16
Last commit
5 days ago

What is it?

LangChain.dart is an unofficial Dart port of the LangChain Python framework. It gives Dart and Flutter developers ready-made components for working with language models, along with a common interface for chaining them into larger workflows such as chatbots, retrieval-augmented generation (RAG), and tool-using agents.

The library is split into modular packages. langchain_core holds the base abstractions and LangChain Expression Language (LCEL), langchain adds higher-level chains, agents and retrieval algorithms, and integration packages such as langchain_openai, langchain_google, langchain_ollama, langchain_anthropic and langchain_mistralai connect to specific providers. Vector store integrations cover Chroma, Pinecone and Supabase, and an in-memory vector store is included for prototyping.

Because it is written in Dart, the same code can run in Flutter mobile, desktop and web apps as well as server-side Dart. Separate API client packages, such as openai_dart and ollama_dart, are also published for developers who only need direct access to a provider's API.

Key features

  • 01Unified interface for chat models from OpenAI, Google Gemini, Anthropic, Mistral and Ollama.
  • 02Composable pipelines via LangChain Expression Language (LCEL).
  • 03Retrieval support with document loaders, embeddings, vector stores and retrievers for RAG.
  • 04Tool-calling agents and built-in tools such as a calculator and Tavily search.
  • 05Works in Flutter apps as well as server-side Dart.

Use cases

  • Adding a chatbot or assistant to a Flutter mobile or desktop app.
  • Building a RAG question-answering feature over your own documents in Dart.
  • Prototyping agents that call external tools from a Dart backend.
  • Switching between LLM providers without rewriting application logic.

Quick start

  1. 1

    Add the langchain package from pub.dev to your Dart or Flutter project

Things to know

  • This is an unofficial Dart port of the Python LangChain framework, not an official LangChain project.

In collections

Open-source AI agent frameworks & apps

Topics

#llm#dart#flutter#rag#langchain#agents#vector-database#ai#generative-ai#llms

FAQ

What is langchain_dart?
A Dart and Flutter port of LangChain for building LLM-powered apps with pluggable model providers, RAG, and agents. LangChain.dart is an unofficial Dart port of the LangChain Python framework. It gives Dart and Flutter developers ready-made components for working with language models, along with a common interface for chaining them into larger workflows such as chatbots, retrieval-augmented generation (RAG), and tool-using agents.
Is langchain_dart free to use? What license does it have?
langchain_dart is open source under the MIT license. Check the license terms before using it in commercial products.
Is langchain_dart still maintained?
Actively maintained. The last commit to langchain_dart was on Nov 2, 2026.
What can I use langchain_dart for?
Adding a chatbot or assistant to a Flutter mobile or desktop app. Building a RAG question-answering feature over your own documents in Dart. Prototyping agents that call external tools from a Dart backend. Switching between LLM providers without rewriting application logic.
How do I install langchain_dart?
Add the langchain package from pub.dev to your Dart or Flutter project
Where can I get langchain_dart?
The source code and installation instructions are on GitHub: https://github.com/davidmigloz/langchain_dart

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects