TEST #3056 RESULTS
03/11/2026 - 8:43 PM
ACCELERATOR
MODEL
24.1
tokens/s
generation
4.02
sec
time to first token
324
tokens/s
prompt
125
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen3 14B - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 5800X3D 8-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
385
tokens/s
28.4
tokens/s
2.70
sec
pp4096+tg256
299
tokens/s
13.4
tokens/s
13.79
sec
pp2048+tg256
352
tokens/s
21.1
tokens/s
5.87
sec
pp2048+tg768
349
tokens/s
20.4
tokens/s
5.92
sec
pp1024+tg1024
373
tokens/s
24.8
tokens/s
2.78
sec
pp1280+tg3072
357
tokens/s
18.7
tokens/s
3.63
sec
pp384+tg1152
391
tokens/s
28.8
tokens/s
1.01
sec
pp64+tg1024
155
tokens/s
30.8
tokens/s
438
ms
pp16+tg1536
258
tokens/s
30.3
tokens/s
91
ms