docs/tutorials/

[giscus[bot]]

docs/tutorials/

New to LangChain or LLM app development in general? Read this material to quickly get up and running building your first applications.

https://python.langchain.com/docs/tutorials/

[SuperMissKuang]

Thanks for reordering the tutorials! It makes more sense to me going through the semantic searching tutorial before building a chatbot using Langchain.

[ZcyhFromC]

The tutorials really do make sense for a newbie. Thanks from the bottom of my heart.

[teddylee777]

Hello, everyone!

I am Teddy. I am leading a passionate LangChain community from South Korea dedicated to exploring and sharing the potential of LangChain.

Currently, we are working on creating a tutorial to make LangChain more accessible for developers in Global. Inspired by the official LangChain Tutorial, we aim to complete this project within a 6-week timeframe. Our dedication is evident, with over 600 commits each week reflecting the team's enthusiasm and hard work.

Although English is not our first language, we hope that this tutorial will be a helpful resource for developers worldwide who are working with LangChain. We humbly share the project link below, hoping it contributes positively to the community.

The project is still ongoing, and we are working tirelessly to complete it by the end of February 2025. We sincerely appreciate your interest and support!

Project "LangChain-OpenTutorial"
https://github.com/LangChain-OpenTutorial/LangChain-OpenTutorial

Thank you! 😊

[GGBond-hxx]

Failed to get info from https://api.smith.langchain.com
how can i solve this question
please help me

[hesreallyhim]

Hey, I really love these tutorials, they are excellent! My question is - what is the intended way for users to run the tutorials as Jupyter notebooks? I just cloned the GitHub repo and am running them on my local machine, which works fine, but I noticed that this aspect is not documented anywhere on the site, as far as I can tell. I think it would be nice to address this: (a) provide instructions (e.g. "clone the GH repo; navigate to..."); (b) provide a link to open the notebook in Colab; (c) run the notebook with Binder; etc. To be clear, I'm not asking for help, I just think this would be a great enhancement to make the notebooks more accessible. I'll probably follow up with an issue on GitHub as well. Thanks!

[hesreallyhim]

OK, my mistake, my PrivacyBadger extension was blocking the Colab badge from showing up(!)

Still, it seems like there is some mixture of concerns going on here, in terms of this page trying to be both a notebook and a React site - for example, even if I open this in Colab, there is still React code visible in the notebook, e.g.:

First up, let's learn how to use a language model by itself. LangChain supports many different language models that you can use interchangeably. For details on getting started with a specific model, refer to supported integrations.

import ChatModelTabs from "@theme/ChatModelTabs";

<ChatModelTabs overrideParams={{openai: {model: "gpt-4o-mini"}}} />

IMO, this is at least distracting, and potentially confusing if a user reads the React import ChatModelTabs... line as relevant to running the notebook. Maybe you can provide links to "pure" .ipynb notebooks that don't have any (non-specially-handled) React code? Or perhaps replace the React tabs with an ipython-based alternative? I will try to submit a PR with an alternative implementation that doesn't result in displaying raw .jsx code when viewing these notebooks in a jupyter notebook environment.

[G4EVA-dev]

First time using LangChain

[ye929]

from langchain.document_loaders import PyPDFLoader
loader = PyPDFLoader("非遗保护政策.pdf")
pages = loader.load_and_split()