TEST #4272 RESULTS
10/12/2026 - 12:13 AM
ACCELERATOR
MODEL
21.4
tokens/s
generation
3.86
sec
time to first token
366
tokens/s
prompt
127
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 5700G with Radeon Graphics (znver3)
RAM
46.9GB
OS
Linux
Kernel Release
7.2.9-2-cachyos
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Fri, 09 Oct 2026 15:36:46 +0000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
410
tokens/s
26.6
tokens/s
2.53
sec
pp4096+tg256
303
tokens/s
9.5
tokens/s
13.63
sec
pp2048+tg256
359
tokens/s
17.9
tokens/s
5.75
sec
pp2048+tg768
359
tokens/s
16.4
tokens/s
5.76
sec
pp1024+tg1024
403
tokens/s
23.0
tokens/s
2.58
sec
pp1280+tg3072
396
tokens/s
14.3
tokens/s
3.27
sec
pp384+tg1152
438
tokens/s
27.0
tokens/s
910
ms
pp64+tg1024
384
tokens/s
29.9
tokens/s
197
ms
pp16+tg1536
246
tokens/s
28.2
tokens/s
95
ms