TEST #2398 RESULTS
01/30/2026 - 3:30 PM
ACCELERATOR
MODEL
135
tokens/s
generation
120
ms
time to first token
13054
tokens/s
prompt
2448
LocalScore
HOW YOU STACK UP
Explore All ResultsSmolDocling 256M Preview - 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
17708
tokens/s
119
tokens/s
64
ms
pp4096+tg256
8326
tokens/s
134
tokens/s
500
ms
pp2048+tg256
13490
tokens/s
135
tokens/s
159
ms
pp2048+tg768
13597
tokens/s
136
tokens/s
157
ms
pp1024+tg1024
17342
tokens/s
134
tokens/s
66
ms
pp1280+tg3072
17630
tokens/s
135
tokens/s
79
ms
pp384+tg1152
22321
tokens/s
134
tokens/s
23
ms
pp64+tg1024
5852
tokens/s
139
tokens/s
17
ms
pp16+tg1536
1219
tokens/s
151
tokens/s
19
ms