TEST #4112 RESULTS
09/01/2026 - 8:36 AM
ACCELERATOR
MODEL
37.2
tokens/s
generation
430
ms
time to first token
3037
tokens/s
prompt
640
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen Threadripper 9960X 24-Cores
RAM
125.2GB
OS
Linux
Kernel Release
6.8.0-136-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #136-Ubuntu SMP PREEMPT_DYNAMIC Wed Jul 1 21:53:05 UTC 2026
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
3341
tokens/s
41.4
tokens/s
330
ms
pp4096+tg256
2875
tokens/s
23.0
tokens/s
1.47
sec
pp2048+tg256
3595
tokens/s
31.5
tokens/s
601
ms
pp2048+tg768
3665
tokens/s
30.4
tokens/s
590
ms
pp1024+tg1024
4318
tokens/s
37.2
tokens/s
261
ms
pp1280+tg3072
3433
tokens/s
29.5
tokens/s
398
ms
pp384+tg1152
4151
tokens/s
45.1
tokens/s
111
ms
pp64+tg1024
1395
tokens/s
49.7
tokens/s
65
ms
pp16+tg1536
556
tokens/s
46.7
tokens/s
47
ms