TEST #2708 RESULTS
02/17/2026 - 11:50 PM
ACCELERATOR
MODEL
646
tokens/s
generation
61
ms
time to first token
21761
tokens/s
prompt
6141
LocalScore
HOW YOU STACK UP
Explore All ResultsTinyStories 656K - Q2_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
28282
tokens/s
754
tokens/s
37
ms
pp4096+tg256
30179
tokens/s
662
tokens/s
138
ms
pp2048+tg256
30501
tokens/s
728
tokens/s
69
ms
pp2048+tg768
21800
tokens/s
676
tokens/s
96
ms
pp1024+tg1024
28154
tokens/s
716
tokens/s
40
ms
pp1280+tg3072
14961
tokens/s
682
tokens/s
87
ms
pp384+tg1152
39031
tokens/s
577
tokens/s
17
ms
pp64+tg1024
2299
tokens/s
557
tokens/s
31
ms
pp16+tg1536
642
tokens/s
461
tokens/s
31
ms