TEST #4113 RESULTS
09/01/2026 - 9:35 PM
ACCELERATOR
MODEL
129
tokens/s
generation
2.20
sec
time to first token
710
tokens/s
prompt
347
LocalScore
HOW YOU STACK UP
Explore All Resultsqwen2.5-0.5b-instruct - Q4_0
SYSTEM
CPU
AMD Ryzen AI 7 PRO 350 w/ Radeon 860M
RAM
23.1GB
OS
Linux
Kernel Release
6.17.0-1032-oem
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #32-Ubuntu SMP PREEMPT_DYNAMIC Fri Jul 24 05:46:26 UTC 2026
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
683
tokens/s
114
tokens/s
1.51
sec
pp4096+tg256
532
tokens/s
107
tokens/s
7.71
sec
pp2048+tg256
625
tokens/s
124
tokens/s
3.29
sec
pp2048+tg768
630
tokens/s
125
tokens/s
3.26
sec
pp1024+tg1024
702
tokens/s
131
tokens/s
1.47
sec
pp1280+tg3072
665
tokens/s
122
tokens/s
1.93
sec
pp384+tg1152
794
tokens/s
146
tokens/s
492
ms
pp64+tg1024
918
tokens/s
146
tokens/s
76
ms
pp16+tg1536
840
tokens/s
146
tokens/s
26
ms