TEST #3592 RESULTS
05/21/2026 - 6:16 PM
ACCELERATOR
MODEL
48.1
tokens/s
generation
675
ms
time to first token
1916
tokens/s
prompt
515
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 5950X 16-Core Processor (znver3)
RAM
31.9GB
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
2297
tokens/s
50.5
tokens/s
467
ms
pp4096+tg256
1896
tokens/s
44.4
tokens/s
2.19
sec
pp2048+tg256
2141
tokens/s
47.9
tokens/s
978
ms
pp2048+tg768
2150
tokens/s
47.3
tokens/s
974
ms
pp1024+tg1024
2277
tokens/s
48.7
tokens/s
471
ms
pp1280+tg3072
2216
tokens/s
47.0
tokens/s
598
ms
pp384+tg1152
2307
tokens/s
49.3
tokens/s
280
ms
pp64+tg1024
1504
tokens/s
49.2
tokens/s
63
ms
pp16+tg1536
461
tokens/s
49.1
tokens/s
54
ms