TEST #3235 RESULTS
04/05/2026 - 6:28 AM
ACCELERATOR
MODEL
3.1
tokens/s
generation
13.29
sec
time to first token
99
tokens/s
prompt
28
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 8700G w/ Radeon 780M Graphics (znver4)
RAM
31.5GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
111
tokens/s
3.4
tokens/s
9.48
sec
pp4096+tg256
97
tokens/s
0.6
tokens/s
43.73
sec
pp2048+tg256
106
tokens/s
2.4
tokens/s
19.75
sec
pp2048+tg768
106
tokens/s
2.1
tokens/s
19.75
sec
pp1024+tg1024
112
tokens/s
3.0
tokens/s
9.47
sec
pp1280+tg3072
108
tokens/s
1.5
tokens/s
12.17
sec
pp384+tg1152
114
tokens/s
4.2
tokens/s
3.55
sec
pp64+tg1024
49
tokens/s
5.5
tokens/s
1.43
sec
pp16+tg1536
89
tokens/s
4.7
tokens/s
288
ms