I’ve installed koboldcpp on a thinkpad x1 with 32gb RAM and a i7-1355U, no GPU. Sure, it’s only just around 1 token/s but for a chat it is still usable (about 15 s per reply). The setup was easier than expected.

  • @magn418MEnglish
    1•3 days ago

    I think doing the calculations at full precision (FP16) is a waste. You should try somewhere between the Q4_K_M version to Q6_K (or at least Q8_0, that’s supposed to the same quality as FP16). That way it should be considerably faster… At least twice as fast.

    (The GGUF page of that model has a list of recommended quantization levels.)

    • @raffaOPEnglish
      2•3 days ago

      thanks for the tips!