bestmodel.run / model — Ornith-1.5-9B-GGUF (agent view)

category chat · ornith-ai/Ornith-1.5-9B-GGUF · 52 runs in the model record
answer: 41.7 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
  Arc Pro B70 32GB                   | 8-bit | 49.6 tok/s     | reported | ttft -ms | vram - GB | ctx 8,192
  BC-250 8GB                         | 4-bit | 42.1 tok/s     | measured | ttft 12,194ms | vram - GB | ctx 12,554
  BC-250 8GB                         | 8-bit | 34.1 tok/s     | measured | ttft 10,415ms | vram - GB | ctx 12,553
  RTX 3060 12GB                      | 4-bit | 65 tok/s       | reported | ttft 60ms | vram - GB | ctx 65,536
  RTX 3080 12GB                      | 5-bit | 92 tok/s       | reported | ttft 762ms | vram 9.5 GB | ctx 262,144
  RTX 3080 12GB                      | 6-bit | 82.8 tok/s     | reported | ttft 800ms | vram 8.5 GB | ctx 131,072