I’ve been using llama.cpp, gpt-llama and chatbot-ui for a while now, and I’m very happy with it. However, I’m now looking into a more stable setup using only GPU. Is this llama.cpp still still a good candidate for that?

  • Hudsonius@lemmy.ml
    link
    fedilink
    English
    arrow-up
    3
    ·
    1 year ago

    GPTQ-for-llama with ooba booga works pretty well. I’m not sure to what extent it uses CPU, but my GPU is at 100% during inference so it seems to be mainly that.

    • bia@lemmy.mlOP
      link
      fedilink
      English
      arrow-up
      1
      ·
      1 year ago

      I’ve looked at that before. Do you use it with any UI?

      • Hudsonius@lemmy.ml
        link
        fedilink
        English
        arrow-up
        3
        ·
        1 year ago

        Yea it’s called Text Generation web UI. If you check out the Ooba Booga git, it goes into good details. From what I can tell it’s based on the automatic1111 UI for stable diffusion.

        • dragonfyre13
          link
          fedilink
          English
          arrow-up
          2
          ·
          1 year ago

          It’s using Gradio, which is what auto1111 also uses. Both of these are pretty heavy modifications/extensions that do a lot to push Gradio to it’s limits, but that’s package being used in both. Note, it also has an api (checkout the --api flag I believe), and depending on what you want to do there’s various UIs that can hook into the Text Gen Web UI (oobabooga) API in various ways.

      • Equality_for_apples
        link
        fedilink
        English
        arrow-up
        1
        ·
        1 year ago

        Personally, I have nothing but issues with Oogas ui, so I connect Silly Tavern to it or KoboldCPP. Works great

  • gh0stcassette@lemmy.world
    link
    fedilink
    English
    arrow-up
    1
    ·
    edit-2
    1 year ago

    Llama.cpp recently added CUDA acceleration for generation (previously only ingesting the prompt was GPU accelerated), and in my experience it’s faster than GPTQ unless you can fit absolutely 100% of the model in VRAM. If literally a single layer is CPU offloaded, the performance in GPTQ immediately becomes like 30-40% worse than an equivalent CPU offload with llama.cpp

    • bia@lemmy.mlOP
      link
      fedilink
      English
      arrow-up
      0
      ·
      1 year ago

      Haven’t been able to test that out, but saw the change. Particularly interesting for my use case.

      • gh0stcassette@lemmy.world
        link
        fedilink
        English
        arrow-up
        1
        ·
        edit-2
        1 year ago

        What use case would that be?

        I can get like 8 tokens/s running 13b models in q_3_k_L quantization on my laptop, about 2.2 for 33b, and 1.5 for 65b (I bought 64gb of RAM to be able to run larger models lol). 7B was STUPID fast because the entire model fits inside my (8gb) GPU, but 7b models mostly suck (wizard-vicuna-uncensored is decent, every other one I’ve tried was Not).