TEST #4271 RESULTS
10/11/2026 - 11:40 PM
ACCELERATOR
MODEL
4.5
tokens/s
generation
46.77
sec
time to first token
30
tokens/s
prompt
14
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 5700G with Radeon Graphics (znver3)
RAM
46.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
31
tokens/s
4.8
tokens/s
33.46
sec
pp4096+tg256
26
tokens/s
4.0
tokens/s
155.04
sec
pp2048+tg256
29
tokens/s
4.5
tokens/s
70.32
sec
pp2048+tg768
29
tokens/s
4.5
tokens/s
70.43
sec
pp1024+tg1024
31
tokens/s
4.6
tokens/s
33.70
sec
pp1280+tg3072
30
tokens/s
4.3
tokens/s
42.65
sec
pp384+tg1152
32
tokens/s
4.7
tokens/s
12.20
sec
pp64+tg1024
29
tokens/s
4.8
tokens/s
2.40
sec
pp16+tg1536
29
tokens/s
4.8
tokens/s
752
ms