

Langchain-Chatchat
β Text source-checked Nov 18, 2026Local knowledge-based RAG and Agent application built with LangChain for LLMs like ChatGLM, Qwen, and Llama.
Source-checked against the vendor's site Β· rebuilt from the vendor's own pagesProfile updated Link checked How we rate
In short
Langchain-Chatchat is a free, open-source RAG and Agent app for offline deployment. It supports local LLMs like GLM-4 and Llama 3 via frameworks such as Ollama and Xinference.
Langchain-Chatchat is an open-source RAG and Agent application project implemented based on large language models such as ChatGLM and application frameworks such as LangChain, designed for offline deployment. The latest version of Langchain-Chatchat allows connecting to models like GLM-4-Chat, Qwen2-Instruct, and Llama3 via frameworks such as Xinference and Ollama, supporting API calls through FastAPI or operation via a Streamlit-based WebUI. Langchain-Chatchat supports mainstream open-source LLMs, Embedding models, and vector databases, enabling fully offline private deployment using open-source models.
The implementation process of Langchain-Chatchat includes loading files, reading text, splitting text, vectorizing text, vectorizing questions, matching the top k most similar text vectors, adding the matched text as context to the prompt, and submitting it to the LLM to generate an answer. The core functionality of the 0.3.x version of Langchain-Chatchat is implemented by the Agent, though users can also manually implement tool calls.
Langchain-Chatchat supports usage in Python 3.8-3.11 environments and has been tested on Windows, macOS, and Linux operating systems. Regarding hardware, the 0.3.0 version of Langchain-Chatchat supports different model deployment frameworks, allowing usage under various hardware conditions such as CPU, GPU, NPU, and MPS.
The code of Langchain-Chatchat follows the Apache-2.0 license.
According to Langchain-Chatchat: Langchain-Chatchat is an open-source RAG and Agent application project implemented based on large language models such as ChatGLM and application frameworks such as LangChain, Designed for offline deployment. Its pages list this strength: Open-source and offline-deployable RAG and Agent application.
β The AI Tool Bible editorial team
Pros
- β Open-source and offline-deployable RAG and Agent application.
- β Supports mainstream open-source LLMs, Embedding models, and vector databases.
- β Provides API call services based on FastAPI and a Streamlit-based WebUI for operation.
- β Usable on various hardware conditions including CPU, GPU, NPU, and MPS.
Facts
- Platforms
- api (not source-checked)
- API available
- Yes (not source-checked)
- Open source
- Yes (license: Apache-2.0)
- GitHub stars
- 38,715 (checked Nov 13, 2026) (not source-checked)
- Last GitHub push
- Nov 10, 2025 (not source-checked)
- First commit
- Mar 2023 (not source-checked)
- Launched
- Jan 1, 2023 (not source-checked)
- Model used
- Multi-model (GLM-4, Qwen2, Llama 3, etc. via Xinference/Ollama/LocalAI/FastChat) (not source-checked)
- Package downloads (30 d)
- 454 Β· pypi langchain-chatchat Β· checked Nov 14, 2026 (not source-checked)
Where this tool sits
Use cases
Frequently asked
- Is Langchain-Chatchat free to use?
- Yes, Langchain-Chatchat is free and open-source. The code follows the Apache-2.0 license, allowing you to deploy it privately without licensing fees.
- Which LLMs does Langchain-Chatchat support?
- It supports mainstream open-source LLMs including GLM-4-Chat, Qwen2-Instruct, and Llama3. You can connect these models via frameworks like Xinference, Ollama, LocalAI, or FastChat.
- Can I run Langchain-Chatchat offline?
- Yes, it is designed for offline private deployment. It enables fully offline usage by supporting open-source models, embedding models, and vector databases without requiring external cloud services.
- What hardware is required for deployment?
- The 0.3.0 version supports various hardware conditions, including CPU, GPU, NPU, and MPS. It works on Windows, macOS, and Linux operating systems with Python 3.8-3.11.
- How does the RAG process work in Langchain-Chatchat?
- It loads files, splits and vectorizes text, matches top-k similar vectors to user questions, adds matched text as context to the prompt, and submits it to the LLM for an answer.