Real time RAG App using Llama 3.2 and Open Source Stack on CPU

Поділитися
Вставка
  • Опубліковано 15 січ 2025

КОМЕНТАРІ • 48

  • @AbhijeetSingh-ku4bz
    @AbhijeetSingh-ku4bz 3 місяці тому +3

    This is great bro. I have been trying to build this for over 2 months and just could progress. This is awesome

    • @AIAnytime
      @AIAnytime  3 місяці тому

      Glad to hear that...

  • @areefa6268
    @areefa6268 День тому

    Hey such a great tutorial. This project is a go to for handson RAG. Thanks. Can you please make a video for hosting these streamlit projects also running llm through ollama. Curious fto know this how you host the llm behind and the proj.

  • @educationdelightenglish3819
    @educationdelightenglish3819 3 місяці тому

    Boss you are a GENIUS! thanks buddy !

  • @avinashnair5064
    @avinashnair5064 3 місяці тому +4

    how can we add convesational memeory buffer i need to have context in my chat who can we do that?

    • @ritikkumar30
      @ritikkumar30 2 місяці тому

      I have same problem of storing context for each session

  • @kareemyoussef2304
    @kareemyoussef2304 3 місяці тому +2

    What about bulk processing the pdfs

  • @mahomaesdios11
    @mahomaesdios11 Місяць тому

    This is great content! Awesome!
    Could I ask you to record a video using Open WebUI with ollama for example? I think WebUI is top-notch.

    • @AIAnytime
      @AIAnytime  Місяць тому

      I already have a lot of videos on that

  • @manikumar-vr3kp
    @manikumar-vr3kp 6 днів тому

    why did u show its perfomance and accuracy

  • @RedCloudServices
    @RedCloudServices 3 місяці тому +1

    but multimodal should mean you can use a Vision LLM to read each PDF page as an image not use traditional vector based RAG - can you make a video using Vision LLM with this PDF?

  • @parwindersingh2302
    @parwindersingh2302 2 місяці тому

    Brilliant bro..better than the so called experts..end to end flow awesome..can we do it for json messages also instead of pdfs. If yes what would be the changes

  • @pedrointeraminense4890
    @pedrointeraminense4890 9 днів тому

    I create the Embbedings, but it generates an error in the "chat with document”, when I ask the question it generates this error: "An error occurred while processing your request: [WinError 10061]". Can anyone help?

  • @muhammedajmalg6426
    @muhammedajmalg6426 3 місяці тому

    thanks for sharing

  • @muhammedaslama9908
    @muhammedaslama9908 3 місяці тому +1

    If Qdrant is currently running locally, what steps should we take when deploying the application?

    • @AIAnytime
      @AIAnytime  3 місяці тому

      You can deploy the same qdrant on any VM via docker command or within a container instance services.

  • @artur50
    @artur50 3 місяці тому

    great! might be an option to add an option for 'semantic search'

  • @stevencheney4159
    @stevencheney4159 3 місяці тому

    Great video - tons of useful information here so thank you!
    Are you able to create embeddings for document types other than PDFs easily? For example, I have a lot of documentation that is stored with Markdown. I could probably export them to a PDF but I'm curious if I could parse them directly.

  • @avinashnair5064
    @avinashnair5064 3 місяці тому

    What is i need to retrive the image from the documents not generation just the images from the pdf with context similarity?

  • @KINGOFCODES
    @KINGOFCODES 2 місяці тому

    Nice....Make End to End Project on TAG....Tabular Augmented Retrieval .....Thanks

    • @AIAnytime
      @AIAnytime  2 місяці тому +1

      I'll add it to my list.

  • @rahulhalappa3927
    @rahulhalappa3927 3 місяці тому

    does it process images within pdf ?

  • @IsmailIfakir
    @IsmailIfakir 3 місяці тому

    is there is a multimodal llm can fine-tuning for sentiment analysis from text, image, video and audio ?

  • @mohsenghafari7652
    @mohsenghafari7652 3 місяці тому

    Thanks, dear. Is it support Persian pdfs and language?

  • @deekshitht786
    @deekshitht786 Місяць тому

    you are awesome 😍

  • @MrRobots100
    @MrRobots100 3 місяці тому

    Can we host this online, to have an API which can react with a client. Can the hugging face 16GB CPU host this with fast interference ? If not, how can I host it on my local device to have an API.

  • @hardiksharma9817
    @hardiksharma9817 3 місяці тому

    how to stop hallucinations in chatbot

  • @maheswarib199
    @maheswarib199 3 місяці тому

    Can anybody tell how different domain of dataset interoperate ?

  • @maheswarib199
    @maheswarib199 3 місяці тому

    Can use hnsw for vector db and train using different domain dataset agriculture legal health for single qa project

  • @light_70
    @light_70 2 місяці тому

    Great Video

  • @saimanikantak-ti2td
    @saimanikantak-ti2td 3 місяці тому

    Could you please share the discord link

  • @SnehaRoy-xf3zv
    @SnehaRoy-xf3zv 3 місяці тому

  • @SonGoku-pc7jl
    @SonGoku-pc7jl Місяць тому

    thanks for all for all for all! sorry for this but, you can make this with mymupd4llm and 3.2 11b with vision please? with next plase!, I all trys conection next with fastapi etc fail :P :(

  • @theindianrover2007
    @theindianrover2007 3 місяці тому

    Awsm

  • @BetterProgrammer-p5k
    @BetterProgrammer-p5k 2 місяці тому +1

    An error occurred while processing your request: [WinError 10061] No connection could be made because the target machine actively refused it
    can someone help me with the above error

    • @sifu1077
      @sifu1077 Місяць тому

      I have the same problem. I think it's firewall related, but I can't figure it out.

    • @pedrointeraminense4890
      @pedrointeraminense4890 9 днів тому

      Is this problem when you get to Chat? Were you able to solve this problem?

  • @rraviteja
    @rraviteja 3 місяці тому +1

    Bro can you please make a video using complete cloud because some people like me don't have good laptop

    • @AIAnytime
      @AIAnytime  3 місяці тому +1

      Already there. Watch my End to end RAG videos using AWS and Azure in RAG Playlist.

    • @mangaanime7727
      @mangaanime7727 3 місяці тому

      Hello Sir,
      Do you provide one on one courses online?

    • @AIAnytime
      @AIAnytime  3 місяці тому

      No

  • @ROKKor-hs8tg
    @ROKKor-hs8tg 10 днів тому

    I want rag pdf without ollama or key
    Llama3.2,chfma,faisscpu,oll freeeeeeee

  • @mohamedfouad1309
    @mohamedfouad1309 Місяць тому

    whats your discord?