vLLM on Kubernetes in Production

แชร์
ฝัง
  • เผยแพร่เมื่อ 12 พ.ย. 2024

ความคิดเห็น • 12

  • @JohnCodes
    @JohnCodes 5 หลายเดือนก่อน +6

    Thanks for having me on Saiyam!! It was alot of fun to show you how we use vLLM at OpenSauced!! Happy to answer any questions here people might have!

    • @DestinoDello
      @DestinoDello หลายเดือนก่อน

      Can share the yaml for deployment please?

  • @aireddy
    @aireddy 4 หลายเดือนก่อน +1

    This is absolutely wonderful session to understand how can we deploy LLMs in production on Kubernetes cluster!!

    • @kubesimplify
      @kubesimplify  4 หลายเดือนก่อน

      @@aireddy glad it was helpful!

  • @nickytonline
    @nickytonline 2 หลายเดือนก่อน

    Great video and great breakdown @JohnCodes!

  • @DaewonSuh
    @DaewonSuh 3 หลายเดือนก่อน

    Thanks for the wonderful Demo!
    I was wondering why you deploy vllm pod through demonsets rather than deployments.
    With daemonset, you can only deploy one pod in one node and a pod occupying a single gpu.
    Considering that nodes are usually attached with multiple gpus, I am afraid that using daemonset might make a lot of gpus idle.

  • @umeshjaiswal5298
    @umeshjaiswal5298 5 หลายเดือนก่อน

    Thanks for this tutorial Saiyam.

    • @kubesimplify
      @kubesimplify  5 หลายเดือนก่อน

      Glad its useful, you building something with LLM?

  • @divyamchandel8734
    @divyamchandel8734 4 หลายเดือนก่อน

    Hi John / Saiyam. In the last part you mentioned "In lot of cases could be cheaper"
    What are those cases where locally hosting it is cheaper vs when using openai is cheaper:
    Is it just dependent on the load which we will have (RPD and max RPM)?

    • @matrix9083
      @matrix9083 4 หลายเดือนก่อน

      openai is $.50 per million tokens for gpt 3.5 for example. If you rent a gpu server for that same amount, you can generate tens or hundred of millions of tokens in one hour depending on which text generation model you choose. something like mistral 7b, phi 3 series, llama 3 8b, gemma 2b,etc all deliver about the same results if not better than gpt 3.5 and also all fit on a gpu server that costs 44 cents per hour on runpod. (the A5000 gpu server for example.)

  • @shivangsharma1
    @shivangsharma1 3 หลายเดือนก่อน +1

    Loved it...❤

    • @kubesimplify
      @kubesimplify  3 หลายเดือนก่อน

      Glad you found it useful!