TEST #2417 RESULTS
01/31/2026 - 8:57 AM
ACCELERATOR
MODEL
117
tokens/s
generation
234
ms
time to first token
6679
tokens/s
prompt
1495
LocalScore
HOW YOU STACK UP
Explore All ResultsReasonLite 0.6B Turbo - Q4_K - Medium
SYSTEM
CPU
Intel Core i5-14450HX (alderlake)
RAM
15.7GB
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
8475
tokens/s
127
tokens/s
129
ms
pp4096+tg256
4440
tokens/s
101
tokens/s
932
ms
pp2048+tg256
6240
tokens/s
116
tokens/s
338
ms
pp2048+tg768
6398
tokens/s
112
tokens/s
329
ms
pp1024+tg1024
7888
tokens/s
118
tokens/s
139
ms
pp1280+tg3072
8307
tokens/s
111
tokens/s
163
ms
pp384+tg1152
11755
tokens/s
125
tokens/s
40
ms
pp64+tg1024
5347
tokens/s
125
tokens/s
20
ms
pp16+tg1536
1256
tokens/s
121
tokens/s
20
ms