Radeon Pro W7900 48GB llama.cpp b9600: shisa-v2.1-unphi4-14b Q4_K_M prefill
1721.0 tok/s prefill (ctx 0.5k, FA on); decode 50.8 t/s tg128
Fields
id
benchmarks/llamabench-w7900-shisa14b-q4km-ppMetric
prefill_tok_s
Value
1,721
Unit
tok/s prefill (ctx 0.5k, FA on); decode 50.8 t/s tg128
Model
shisa-v2.1-unphi4-14b Q4_K_M
Methodology
llamabench.ai community row, reporter @lhl: llama.cpp b9600, ROCm, FA on, kv f16, 0.5k ctx. n=1.
Measured by
third_party