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

Поділитися
Вставка
  • Опубліковано 18 лис 2024

КОМЕНТАРІ • 37

  • @AbhijeetSingh-ku4bz
    @AbhijeetSingh-ku4bz Місяць 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  Місяць тому

      Glad to hear that...

  • @avinashnair5064
    @avinashnair5064 Місяць тому +4

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

    • @ritikkumar30
      @ritikkumar30 5 днів тому

      I have same problem of storing context for each session

  • @kareemyoussef2304
    @kareemyoussef2304 Місяць тому +2

    What about bulk processing the pdfs

  • @parwindersingh2302
    @parwindersingh2302 19 днів 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

  • @BetterProgrammer-p5k
    @BetterProgrammer-p5k 22 дні тому

    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

  • @KINGOFCODES
    @KINGOFCODES 17 днів тому

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

    • @AIAnytime
      @AIAnytime  16 днів тому +1

      I'll add it to my list.

  • @IsmailIfakir
    @IsmailIfakir Місяць тому

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

  • @muhammedaslama9908
    @muhammedaslama9908 Місяць тому +1

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

    • @AIAnytime
      @AIAnytime  Місяць тому

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

  • @avinashnair5064
    @avinashnair5064 Місяць тому

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

  • @educationdelightenglish3819
    @educationdelightenglish3819 Місяць тому

    Boss you are a GENIUS! thanks buddy !

  • @RedCloudServices
    @RedCloudServices Місяць 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?

  • @artur50
    @artur50 Місяць тому

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

  • @MrRobots100
    @MrRobots100 Місяць 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.

  • @rahulhalappa3927
    @rahulhalappa3927 Місяць тому

    does it process images within pdf ?

  • @maheswarib199
    @maheswarib199 Місяць тому

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

  • @mohsenghafari7652
    @mohsenghafari7652 Місяць тому

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

  • @maheswarib199
    @maheswarib199 Місяць тому

    Can anybody tell how different domain of dataset interoperate ?

  • @light_70
    @light_70 12 днів тому

    Great Video

  • @stevencheney4159
    @stevencheney4159 Місяць 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.

  • @hardiksharma9817
    @hardiksharma9817 Місяць тому

    how to stop hallucinations in chatbot

  • @saimanikantak-ti2td
    @saimanikantak-ti2td Місяць тому

    Could you please share the discord link

  • @muhammedajmalg6426
    @muhammedajmalg6426 Місяць тому

    thanks for sharing

  • @theindianrover2007
    @theindianrover2007 Місяць тому

    Awsm

  • @SnehaRoy-xf3zv
    @SnehaRoy-xf3zv Місяць тому

  • @rraviteja
    @rraviteja Місяць тому +1

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

    • @AIAnytime
      @AIAnytime  Місяць тому +1

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

    • @mangaanime7727
      @mangaanime7727 Місяць тому

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

    • @AIAnytime
      @AIAnytime  Місяць тому

      No