Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

ChatPDF – Chat with Any PDF

chatpdf.com

Post reply on HN
Print view View original

61–70 of 141 posts

Re: ChatPDF – Chat with Any PDF

#61

Post by mediaman »

derwiki wrote: post #27

I uploaded a 750 page novel and asked to summarize the plot, to which it responded: > I'm sorry, but I cannot provide a summary of the plot of this book as the PDF file does not contain any information about the plot. The PDF file only includes the cover design, copyright information, table of contents, and details about the author. It is possible that the book itself contains a summary or synopsis of its plot. Is th…

It is returning a series of snippets from the book, from which it cannot create a summary. That’s because of the way vectorized search works.

Summarizing the book requires a different approach. Usually condensing the book, maybe processing ten pages at a time, and then summarizing the condensed chunks.

Re: ChatPDF – Chat with Any PDF

#62

Post by willseth »

Cool idea, but the LLM was too dumb in my test. I gave it a REST API design book and asked it for a batch update pattern, and it suggested PATCH: "...a good pattern for batch updates is to use the HTTP PATCH method. The PATCH method allows you to update specific fields of a resource, rather than replacing the entire resource. This can be useful for batch updates, as it allows you to update multiple resources with a single request." Oops!

Re: ChatPDF – Chat with Any PDF

#63

Post by rolisz »

derwiki wrote: post #18
bvan wrote: post #6

Great, but I would have to run this in >my< private cloud. no way any business is going to upload its docs into a third-party cloud, no matter what the small print says.

This seems vault-ai based which has instructions to self-host: http://github.com/pashpashpash/vault-ai

Vault still uses Pinecone as a 3rd party service and your embeddings do get sent there.

Re: ChatPDF – Chat with Any PDF

#64

Post by DANmode »

mdrzn wrote: post #29

Posted 4 times in the last 2 months?

It keeps receiving engagement.

Conceptually, it's revolutionary.

In practice, I probably wouldn't have posted more than a couple of times until things like document length restrictions were a thing of the past.

Re: ChatPDF – Chat with Any PDF

#65

Post by rolisz »

bvan wrote: post #6

Great, but I would have to run this in >my< private cloud. no way any business is going to upload its docs into a third-party cloud, no matter what the small print says.

There are various degrees to self hosting: for nice outputs, you need OpenAIs APIs to generate at least the answers. There are alternatives, but not as good.

If you are interested in this, feel free to reach out to me and I can help you with setting this up.

Re: ChatPDF – Chat with Any PDF

#66

Post by roflyear »

lamp987 wrote: post #24
xnx wrote: post #23

4th submission in 3 weeks

from an entirely new account too. jannies should start deleting these ads. ever since gpt-4 dropped, this website has become unbearable.

i'm starting to think it's some conspiracy, reddit is famous for having created fake accounts to fake engagement and activity to get the site off the ground. do we really put this type of stuff beyond the guys at openai? i don't.

Re: ChatPDF – Chat with Any PDF

#67

Post by felipesabino »

sensibar wrote: post #49
felipesabino wrote: post #34

It seems to be a paid version of https://github.com/mayooear/gpt4-pdf-chatbot-langchain It uses langchain and pinecone to create a semantic index over the PDF content and search it based on question asked to sends the relevant information to openAI GPT api using embeddings.

No, it doesn't use langchain and yes it uses OpenAI.

Care to elaborate? I mean, looking at the source code it clearly does use LangChain for PDF ingestion https://github.com/mayooear/gpt4-pdf-chatbot-langchain/blob/...

Re: ChatPDF – Chat with Any PDF

#68

Post by uses »

How does this work? Does it feed the PDF's text as a prompt to the LLM? How would you do this if you had, say, thousands of pages of a website?

I feel like "chatbot/search engine hybrid which can consume a large website and know everything about the org it represents" is a powerful application.

Re: ChatPDF – Chat with Any PDF

#69

Post by ayanb9440 »

bvan wrote: post #6

Great, but I would have to run this in >my< private cloud. no way any business is going to upload its docs into a third-party cloud, no matter what the small print says.

Here is a fully self-hostable solution that connects to PDFs in your google drive folder: https://github.com/ai-sidekick/sidekick

Uses weaviate so that even the vectorstore can be self-hosted

Re: ChatPDF – Chat with Any PDF

#70

Post by ffpip »

felipesabino wrote: post #67
sensibar wrote: post #49

Earlier quoted context omitted.

No, it doesn't use langchain and yes it uses OpenAI.

Care to elaborate? I mean, looking at the source code it clearly does use LangChain for PDF ingestion https://github.com/mayooear/gpt4-pdf-chatbot-langchain/blob/...

He might mean the site chatpdf.com does not use langchain but uses the OpenAI API.
Post reply on HN