--- title: "Discussion on \"How to build a PDF chatbot with Langchain ๐Ÿฆœ๐Ÿ”— and FAISS\" | Hashnode" description: "Discussion on \"How to build a PDF chatbot with Langchain ๐Ÿฆœ๐Ÿ”— and FAISS\". A detailed guide and tutorial on Langchain and FAISS, where I walk your through step by step on how to build your own PDF chatbot." url: "https://hashnode.com/posts/how-to-build-a-pdf-chatbot-with-langchain-and-faiss/65eaa742f5c21d6e4635355b" --- [Feed](https://hashnode.com/feed) / Discussion # [How to build a PDF chatbot with Langchain ๐Ÿฆœ๐Ÿ”— and FAISS](https://kevincoder.co.za/how-to-build-a-pdf-chatbot-with-langchain-and-faiss?utm_source=hashnode&utm_medium=feed) A detailed guide and tutorial on Langchain and FAISS, where I walk your through step by step on how to build your own PDF chatbot. Mar 8, 2024 ยท 8 min read ยท kevincoder.co.za Updated Dec 12, 2024 Read the article - [#machine-learning](https://hashnode.com/tag/machine-learning) - [#python](https://hashnode.com/tag/python) - [#openai-llms-langchain-promttemplate-promptengineering-python](https://hashnode.com/tag/openai-llms-langchain-promttemplate-promptengineering-python) - [#langchain](https://hashnode.com/tag/langchain) - [#vector-embeddings](https://hashnode.com/tag/vector-embeddings) ## Written by [Kevin Naidoo](https://hashnode.com/@kwnaidoo) @kwnaidoo ยท Durban, South Africa [View profile](https://hashnode.com/@kwnaidoo) Passionate & experienced tech enthusiast On kevincoder.co.za, I write about my journey as a developer working across Django, Go, and everything in between, from large-scale systems to small, useful tools. Programming has been my passion for about 2 decades now, and I'm still loving it, so I experiment a lot and sometimes write about it. ## Responses 1 Markdown supported ยท โŒ˜โ†ต to post [Elon Musk](https://hashnode.com/@Elon817) [May 19, 2024](https://hashnode.com/posts/how-to-build-a-pdf-chatbot-with-langchain-and-faiss/65eaa742f5c21d6e4635355b/comment/664a647ca826decae35a6035) Thank you [Kevin Naidoo](https://hashnode.com/@kwnaidoo) for this useful guide. If someone wants to develop a chatbot whose knowledge base is drawn from the content (posts articles, and comments) of a website, how can they dynamically update the `memory` of `chain`? By the way, can you tell me about the pricing of openai API key usage with some examples for using the gpt-3.5-turbo model? Thank you! [Kevin Naidoo](https://hashnode.com/@kwnaidoo) [May 22, 2024](https://hashnode.com/posts/how-to-build-a-pdf-chatbot-with-langchain-and-faiss/65eaa742f5c21d6e4635355b/comment/664a647ca826decae35a6035#reply-664e0a844e5ea0649ba29614) Hi [Elon Musk](https://hashnode.com/@Elon817) . RAG is essentially taking text data and vectorizing the data so that it can be searched. In the case of a website, you need to scrape the text of the website and then follow the RAG process as described in this article. You don't need to always use RAG, if you use Langchain tools for example, you can easily just query an API or scrape a page and then provide that data to the model in real-time. RAG helps to narrow down large chunks of text. Remember you pay for each token and models usually have a limitation on the maximum number of tokens you can send in a prompt, therefore using a RAG system - you minimize the amount of tokens sent. OpenAI uses a token-based costing system. A token is roughly 4 characters (in English). You get billed a different rate for text fed into the model and text generated from the model. GPT3.5-Turbo is generally the most cost-effective model, it's not the best model but for most use cases it's usually fine. With GPT3.5 Turbo expect to pay $0.50 / 1M tokens for text you pass into the model, and $1.50 / 1M tokens the model generates. [Elon Musk](https://hashnode.com/@Elon817) [May 23, 2024](https://hashnode.com/posts/how-to-build-a-pdf-chatbot-with-langchain-and-faiss/65eaa742f5c21d6e4635355b/comment/664a647ca826decae35a6035#reply-664f2e6793310d32b04302a2) Hi [Kevin Naidoo](https://hashnode.com/@kwnaidoo), Thanks for your comprehensive response! I am developing a chatbot for a WordPress website and planning to use your script as the chatbot backend. However, I'm curious about how to implement the asynchronous function effectively. Since the chatbot will be used on a public website, I should consider making multiple calls on the chatbot in a short period. Do you have any recommendations on how to approach this aspect? Thanks, Elon [Kevin Naidoo](https://hashnode.com/@kwnaidoo) [May 23, 2024](https://hashnode.com/posts/how-to-build-a-pdf-chatbot-with-langchain-and-faiss/65eaa742f5c21d6e4635355b/comment/664a647ca826decae35a6035#reply-664f672472017ae1482cc175) [Elon Musk](https://hashnode.com/@Elon817) For chat, generally, web sockets make the most sense. PHP does have web socket support via: [openswoole.com](https://openswoole.com/) Open Swoole might introduce more complexity than you need, depending on how many connections you want to handle per second/minute, might just be better to use regular AJAX and HTTP requests. PHP 8 with PHP-FPM can handle these fairly efficiently, or in the Python world, FastAPI is a good choice. FYI: I do have a few tutorials on this blog relating to FastAPI and also using Qdrant as a vector store instead of FAISS. FAISS may not work so well for more advanced realtime chatbots.