bestmodel.run / model — gemma-4-E2B-it (agent view)

category chat · google/gemma-4-E2B-it · 17 runs in the model record
answer: 106.8 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
  AMD Ryzen 7 8745H w/ Radeon 780M   | 4-bit | 47.4 tok/s     | reported | ttft 485ms | vram 2.5 GB | ctx 32,768
  Intel(R) Core(TM) Ultra 7 155H     | 4-bit | 27.6 tok/s     | reported | ttft -ms | vram - GB | ctx 2,048
  Qualcomm Snapdragon 888 ARM64      | 4-bit | 6.2 tok/s      | reported | ttft 10,000ms | vram 6 GB | ctx 1,306
  GTX 1060 3GB                       | 4-bit | 48.2 tok/s     | reported | ttft 8,399ms | vram - GB | ctx 2,048
  Radeon AI Pro R9700 32GB           | 8-bit | 169.7 tok/s    | reported | ttft -ms | vram - GB | ctx 2,048
  Radeon AI Pro R9700 57GB           | 4-bit | 47.4 tok/s     | reported | ttft 493ms | vram - GB | ctx 2,048
  RTX 3060 12GB                      | 4-bit | 133.4 tok/s    | reported | ttft 112ms | vram 3.8 GB | ctx 4,096
  RX 570 4GB                         | 4-bit | 3.7 tok/s      | reported | ttft 199,403ms | vram - GB | ctx 2,048
  RX 570 4GB ×2                      | 2-bit | 16 tok/s       | reported | ttft 246ms | vram - GB | ctx 2,048
  RX 5700 8GB                        | 4-bit | 73.1 tok/s     | reported | ttft -ms | vram 3 GB | ctx 2,048
  Ryzen AI Max 395 128GB             | 4-bit | 107 tok/s      | measured | ttft -ms | vram 8.5 GB | ctx 32,768
  Ultra 7 155H 32GB                  | 4-bit | 28 tok/s       | reported | ttft -ms | vram - GB | ctx 2,048