Inference · hardware · own machine

What runs on AMD Radeon(TM) Vega 8 Mobile Graphics · 1 GB.

Every open model a machine of this kind has run on the network, at its fastest build, fastest first. Speed is tokens a second, measured and signed by the machine itself.

3models measured
27fastest tok/s
1machine
ModelFastest build heretok/sContext$ / M tokens
Gemma 3 270MQ4_K_M (4-bit) by unsloth27—$6.41
Llama 3.2 1Bbuild5.64K$31.16
MiniCPM-V 4 BaseQ4_K_M (4-bit) by openbmb3.7—$47.40