TEST #2747 RESULTS
02/18/2026 - 12:25 AM
ACCELERATOR
MODEL
35.4
tokens/s
generation
1.59
sec
time to first token
739
tokens/s
prompt
254
LocalScore
HOW YOU STACK UP
Explore All ResultsSmolLM2 135M Instruct - IQ1_S - 1.5625 bpw
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
980
tokens/s
40.5
tokens/s
1.07
sec
pp4096+tg256
874
tokens/s
34.4
tokens/s
4.73
sec
pp2048+tg256
861
tokens/s
38.6
tokens/s
2.41
sec
pp2048+tg768
803
tokens/s
32.9
tokens/s
2.59
sec
pp1024+tg1024
887
tokens/s
33.8
tokens/s
1.18
sec
pp1280+tg3072
865
tokens/s
34.5
tokens/s
1.51
sec
pp384+tg1152
803
tokens/s
35.9
tokens/s
500
ms
pp64+tg1024
435
tokens/s
34.1
tokens/s
174
ms
pp16+tg1536
142
tokens/s
34.3
tokens/s
139
ms