I Just Pulled Someone Out of the DIY LangChain Rabbit Hole! Why go with LangChain+LangGraph in 2024

Подписаться 2,7 тыс.

Просмотров 2,3 тыс.

50% 1

I often get asked whether to implement a framework like ‪@LangChain‬ 🦜️🔗 internally or use the open-source version.
For this, I'm quite clear: Go with LangChain! Here's why:
Opinionated Framework: LangChain, like Flask or Django, has strong opinions on building generative AI applications, incorporating best practices and patterns.
Don't Reinvent the Wheel: Avoid wasting time fixing bugs already solved by others. Leverage the expertise behind LangChain and LangGraph.
Extensibility: LangChain makes it easy to extend and customize with the LangChain Expression Language.
Best Practices + Great Prompt Engineering: LangChain provides refined prompts, offering best practices for generative AI applications.
Community and Expertise: LangChain benefits from thousands of contributors with deep domain knowledge, providing a solid foundation.
Disclaimer: LangChain🦜️🔗 isn't perfect. Some find its abstractions confusing, but the LangChain Expression Language improves transparency. Security is a shared responsibility, so developers must manage it carefully.
I hope you find this video helpful! Share your thoughts, and if you disagree, I'd love to hear your opinions.

Наука

Опубликовано:

13 сен 2024

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 12

@SigAiOC-ke3ss 3 месяца назад

Langchain is moving at such breakneck speed with complete disregard to backwards compatibility that the code you wrote couple months ago is obsolete and is not working anymore... Yes it saves you time when you do a quick test but for production, especially if you care about the ability to upgrade your libraries, I'd always build from scratch.

@IanBicking 3 месяца назад

As an example of something you DON'T need a framework to do: if you want to use multiple models you can do that using any routing service, such as OpenRouter, Martian, or BrainTrust. Not only do they handle the model abstraction (generally making every model look like GPT), but they also handle the billing so you don't need N accounts to support N models. If you start development with GPT but want to try out Claude, Gemini, Mistral, etc., this is the easiest way to go.

@awakenwithoutcoffee 2 месяца назад

hey bro, does LangFlow play a part in your picture or is it more an "'abstraction" programmers should avoid ? great channel btw.

@user-wr4yl7tx3w 3 месяца назад

How about llamaindex?

@bastabey2652 3 месяца назад

gen ai is in too early for frameworks to be opinionated... learn by experimenting with prompts and Python.. don't use black boxes.. if you're a technical developer, these frameworks won't help you anyway I take exception with Llamaindex pdf reader...

@diegocalderon3221 3 месяца назад

LangChain could use some serious library refactoring/organizing. Importing libraries shouldn’t take 40 lines of code.

@EdenMarco 3 месяца назад

can you please elaborate? havn't encountered this myself

@diegocalderon3221 3 месяца назад

@@EdenMarco 40 lines in an exaggeration but not unnormal to have 10-15 lines of code just for imports on any Lang project

@thunkin-ai 2 месяца назад

I study the langchain codebase quite a bit to understand the lessons they're learned and how they've solved them. However, I find langchain to be quite wild and unwieldy and find myself opting to use less and less langchain and more my own abstractions. Langgraph _seems_ be, to me, the approach that Langchain could/should have gone with and I'm finding LG not-too-much-framework.

@jon200y 3 месяца назад

Great videos! keep them coming please.

@AlexanderSomma 3 месяца назад

Why‽ Lang Chain isn't needed if you know how to work with templating, JSON, retrieval, and storage. To be clear, I'm not saying don't use LangCHain. I am saying don't confuse opinionated frameworks for what is right for you. If you like the lying chain approach, go with it for those who have different ideas that are not in line with LangChain or strong opinions. Don't use it; roll your own and share with the community.