TEST #2976 RESULTS
03/05/2026 - 10:57 AM
ACCELERATOR
MODEL
22.3
tokens/s
generation
2.79
sec
time to first token
517
tokens/s
prompt
161
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
63.9GB
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
618
tokens/s
26.6
tokens/s
1.70
sec
pp4096+tg256
391
tokens/s
12.0
tokens/s
10.56
sec
pp2048+tg256
514
tokens/s
19.7
tokens/s
4.03
sec
pp2048+tg768
515
tokens/s
18.7
tokens/s
4.03
sec
pp1024+tg1024
610
tokens/s
22.7
tokens/s
1.72
sec
pp1280+tg3072
585
tokens/s
16.6
tokens/s
2.23
sec
pp384+tg1152
670
tokens/s
26.4
tokens/s
606
ms
pp64+tg1024
467
tokens/s
30.0
tokens/s
167
ms
pp16+tg1536
282
tokens/s
28.4
tokens/s
85
ms