TEST #3153 RESULTS
03/22/2026 - 8:32 PM
ACCELERATOR
MODEL
50.0
tokens/s
generation
656
ms
time to first token
1959
tokens/s
prompt
530
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 5900X 12-Core Processor (znver3)
RAM
127.9GB
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
2358
tokens/s
50.5
tokens/s
454
ms
pp4096+tg256
1906
tokens/s
46.2
tokens/s
2.17
sec
pp2048+tg256
2173
tokens/s
49.7
tokens/s
964
ms
pp2048+tg768
2177
tokens/s
48.7
tokens/s
960
ms
pp1024+tg1024
2352
tokens/s
50.5
tokens/s
455
ms
pp1280+tg3072
2286
tokens/s
49.4
tokens/s
580
ms
pp384+tg1152
2369
tokens/s
52.7
tokens/s
199
ms
pp64+tg1024
1656
tokens/s
50.9
tokens/s
58
ms
pp16+tg1536
349
tokens/s
51.0
tokens/s
66
ms