TEST #2934 RESULTS
02/27/2026 - 5:50 PM
ACCELERATOR
MODEL
11.9
tokens/s
generation
1.16
sec
time to first token
1091
tokens/s
prompt
224
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen3 14B - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 9900X 12-Core Processor
RAM
125.6GB
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
1386
tokens/s
12.4
tokens/s
828
ms
pp4096+tg256
1151
tokens/s
10.9
tokens/s
3.66
sec
pp2048+tg256
1313
tokens/s
12.0
tokens/s
1.64
sec
pp2048+tg768
1313
tokens/s
11.8
tokens/s
1.65
sec
pp1024+tg1024
1401
tokens/s
12.2
tokens/s
814
ms
pp1280+tg3072
1319
tokens/s
11.6
tokens/s
1.06
sec
pp384+tg1152
1384
tokens/s
12.2
tokens/s
386
ms
pp64+tg1024
431
tokens/s
12.2
tokens/s
223
ms
pp16+tg1536
125
tokens/s
12.2
tokens/s
221
ms