Flowise: No Code ChatBot Building Platform: LangChain

[@engineerprompt]

What are you going to create with this tool? 💼Consulting: calendly.com/engineerprompt/consulting-call 🦾 Discord: discord.com/invite/t4eYQRUcXB ☕ Buy me a Coffee: ko-fi.com/promptengineering |🔴 Join Patreon: Patreon.com/PromptEngineering ▶ Subscribe: www.youtube.com/@engineerprompt?sub_confirmation=1

[@DjehutimasAsarRa]

If fails out when I load a large document

[@musumo1908]

What a superb video! Easy to follow - great explanation - 1st flowise test bot built! Thanks

[@engineerprompt]

Glad it was helpful!

[@tilltechdouspart]

Problem Solved Updated the flow from PDF to DOCX Document Ret and it works perfectly. I suspect, the PDF (created by MS Word "save as") isn't the expected or real PDF standard.

[@engineerprompt]

I have also notice that it doesn't work on some PDF files.

[@adamali-lp7pd]

how to use PrivateGPT or gpt4all + langflow or Flowise + save csv to file pls

[@nilsbuer]

Really great! Very good background explenation

[@techwithray8943]

Flowise is a remarkable no-code UI flow builder using JS. While Flowise has made significant strides, it still needs more contribution to the repo...

[@engineerprompt]

Agree :)

[@goforit5]

Great video as usual! So easy to follow and well-explained. Thank you!

[@thanushibasnayake403]

There isn't any language model in conversational retrieval QA . so, How can i connect openAi with that?

[@rodrigofarias900]

besides the UI, whats the difference between this and the Transformer Agents from hugging face, from a recent video of yours?

[@TonyHoangPodcast]

how do you find a chain on flowise that will allow you to connect (memory, document loader, tool)? I can only find a chain that does either (memory/document loader) or (memory/tool)

[@mayorc]

What are the differences between of LocalGPT and PrivateGPT? That side I'm more interested in a flowise tutorial using the local path (hugging face/GPT4All) instead of the openai path.

[@engineerprompt]

By default LocalGPT uses GPU if its available on your machine. PrivateGPT uses CPU only.

[@Betegfos]

Awesome stuff mate! Thanks for the video. How can I use this chatbot commercially on a website?

[@diegosandoval7462]

great work thank you. Do you know if there are any limitations to the size of pdfs I can upload? will that affect the costs? what about the number of pdfs ?

[@jmanhype1]

Also can you make a video on how to incorporate smol dev and flowise

[@jagusiff]

how can you tell whether the Prompt is getting an answer from the Document vs. OpenAI when providing a response?

[@engineerprompt]

You can check the retrieved documents and see if the content is actually there or not

[@barrytrute]

Would it be possible with this model to expand the user interface to support multiple file types easily? Where does the knowledge base reside and can it be encrypted or otherwise secured to protect sensitive data?

[@adsmoberg]

Great video. How to make the bot extract information solely from the PDF file?

[@wordpressmagics2613]

Hello. I have tried it using two objects - Pinecone Load Existing Index and ChatOpenAI model (which has gpt-4 and 3.5-turbo, btw). Test chat shows me an error: this.embeddings.embedQuery is not a function Can you help me to debug it?

[@kishorchannal5805]

Hey, did anyone encounter "Error: An error occurred while fetching the blob"? I tried setting Hugging Face API token to Write, changed the document type to Docx. Somehow it still throws the same error

[@lineupp]

How to get input from webhook and send output via post request? Awesome video

[@denizorsel1029]

Due to Token limits, it is still not a viable robot for production. It can't even consistently answer queries out of a 150kb text file. I m not talking about Flowise but the current AItech unfortunately. There is no other way than training your own data to get consistent results. I hope billion tokens will become a thing soon.

[@tilltechdouspart]

I appreciate your teaching style. Nice. I just followed along and created my first chatflow. but am getting the following error UnknownErrorException: PDFDocument: stream must have data any ideas why? or could you point to where I may be able to get support?

[@engineerprompt]

Thank you 😊 Make sure you are able to copy data from the pdf file. If the pdf has images or the text is not copy able. Then it might not work.

[@adrienkwong]

Thanks. Nice video. What if my data is very larger. For example, 10000 pdfs that are more than 10GB?

[@RajivPerera]

Great video, as always. Can that be a folder instead of one PDF?

[@anubisai]

You can make it ingest many files at a single time. Yes.

[@oliveryang4626]

Instead of simple QnA, how does it work with a bit more complicated work such as taking a detailed note of an one hour meeting in text?

[@engineerprompt]

I haven't tested the summarization capabilities of this no-code tool for langchain has support for summarization models.

[@JustStart562]

i keep getting an insuficient quota error

[@luciaojos]

Would be great if you create the same but with a free LLM :)

[@jmanhype1]

Can you show how to create a discord bot and how to use api calls and webhooks

[@AnhTuấnNguyễn-g3z]

I follow you but I get this error Error: Error: Error: ENOENT: no such file or directory, open ''. Can you help me fix it?

[@sylestra]

did you find out why, I'm getting the same?

[@BrutalStrike2]

9:56

[@Desi_manga_mania]

bruh coding kar bakchodi nai