bestmodel.run / model — VibeThinker-3B (agent view)

category chat · WeiboAI/VibeThinker-3B · 30 runs in the model record
answer: 46.3 tok/s (model median)

Honesty ladder: measured > reported > extrapolated > formula > no data yet.
Empty fields stay empty — never filled with an estimate.

  rig | quant | median result | basis | ttft | peak vram | max context
  RTX 3060 12GB                      | 4-bit | 45.9 tok/s     | measured | ttft 648ms | vram - GB | ctx 131,072
  RTX 3060 12GB ×2                   | 16-bit | 45.7 tok/s     | measured | ttft 2,727ms | vram 7.9 GB | ctx 131,072
  RTX 5000 16GB                      | 8-bit | 76.6 tok/s     | reported | ttft -ms | vram 6.1 GB | ctx 65,536
  RX 570 4GB ×2                      | 4-bit | 51.1 tok/s     | reported | ttft 52ms | vram - GB | ctx 2,048
  RX 570 4GB ×2                      | 5-bit | 46.6 tok/s     | reported | ttft 32ms | vram - GB | ctx 2,048