Nomic's New Embedding Model | nomic-embed-text

Поділитися
Вставка
  • Опубліковано 22 січ 2025

КОМЕНТАРІ • 15

  • @SaiBharadwajKakunuri
    @SaiBharadwajKakunuri 10 місяців тому +2

    I gave a like for expression you mentioned to show how fast nomic embedding is.nice video

  • @limitless1692
    @limitless1692 Місяць тому

    You explained something really important here.
    First that Llama2 model can do vector embeding RAG
    And you showed that nomic-embed-text can do the same RAG embeding, but nomic-embed-text it is alot better and it definetely it is alot faster.
    I mean the speed differance is insane!! Thank you for showing that.

  • @VenkatesanVenkat-fd4hg
    @VenkatesanVenkat-fd4hg 10 місяців тому +2

    Always superb and awesome work....continue. I think bge embeddings also good based on my survey but not by my own experiment...😅

    • @datasciencebasics
      @datasciencebasics  10 місяців тому +1

      yes that is also good. More and more better open embeddings evolving, which is good for us to experiment.

  • @kanishkayohan
    @kanishkayohan 6 місяців тому

    hi,
    1. how can i use embeddings = OllamaEmbeddings(model="nomic-embed-text") model directly from hugging face. Not from locally installed instance ?
    2. if i am using local installed instance how can i publish it to huggingface space

  • @aireddy
    @aireddy 7 місяців тому

    Very detailed explanation

  • @Jeganbaskaran
    @Jeganbaskaran 10 місяців тому

    Thank you for the video. Good to know the new embedding. Did you encounter any accuracy while we were doing the similarity search? is it providing accurate results as BG and other embeddings?

  • @chiraggbharadwajdeptofcomputer
    @chiraggbharadwajdeptofcomputer 9 місяців тому

    hello sir this is a very helpful video but can you please tell us how you got got 3.5 turbo on Ollama list like where did you download the model from

    • @datasciencebasics
      @datasciencebasics  9 місяців тому

      Thank you. In some cases, you might need the naming as gpt 3.5 turbo model in your codebase, hence I was just copying any existing open source model with that name!! Under the hood, its OS model but different naming :)
      View doc here -> github.com/ollama/ollama

  • @SantK1208
    @SantK1208 10 місяців тому

    Sir, please create a playlist so that we can learn easily from your previous videos as well

  • @Maisonier
    @Maisonier 9 місяців тому

    What about LM Studio with Anything LLM? today I have an update for LM Studio with "nomic embeded text" when I start the server. But I always used Anything LLM connected to this server, to chat. But I don't understand all this, how it's working? liked and subscribed.

  • @drmetroyt
    @drmetroyt 10 місяців тому

    Please make a video on anything llm. Project

  • @drmetroyt
    @drmetroyt 10 місяців тому

    Arent all embedding models same ? They do the same thing ? Change my thoughts please

    • @datasciencebasics
      @datasciencebasics  10 місяців тому

      What they do under the hood is same but quality can be different. Hence, choosing good quality embedding model is essential for retrieving better result.