3 ms·

try here, I hate llms but this is crazy fast. https://chatjimmy.ai/ https://chatjimmy.ai/

• bmacho 8mo ago
  "447 / 6144 tokens" "Generated in 0.026s • 15,718 tok/s" This is crazy fast. I always predicted this speed in ~2 years in the future, but it's here, now.
• Lalabadie 8mo ago
  The full answer pops in milliseconds, it's impressive and feels like a completely different technology just by foregoing the need to stream the output.
• FergusArgyll 8mo ago
  Because most models today generate slowish, they give the impression of someone typing on the other end. This is just <enter> -> wall of text. Wild
• machiaweliczny 8mo ago
  We need that for this chinese 3B model that think 45s for hello world but also solves math.
  • Bolwin 8mo ago
    Nanbeige. Yeah this seems ideal for models that scale test time compute
• Serenacula 8mo ago
  Do we know anything about the method?