TEST #3292 RESULTS
04/11/2026 - 10:03 PM
ACCELERATOR
MODEL
12.6
tokens/s
generation
770
ms
time to first token
1638
tokens/s
prompt
299
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 7800X3D 8-Core Processor (znver4)
RAM
31.1GB
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
2126
tokens/s
13.2
tokens/s
558
ms
pp4096+tg256
1941
tokens/s
12.5
tokens/s
2.19
sec
pp2048+tg256
2046
tokens/s
12.7
tokens/s
1.08
sec
pp2048+tg768
2071
tokens/s
12.6
tokens/s
1.07
sec
pp1024+tg1024
2126
tokens/s
12.6
tokens/s
560
ms
pp1280+tg3072
1903
tokens/s
12.7
tokens/s
753
ms
pp384+tg1152
1915
tokens/s
12.6
tokens/s
293
ms
pp64+tg1024
486
tokens/s
12.3
tokens/s
209
ms
pp16+tg1536
126
tokens/s
12.4
tokens/s
217
ms