TEST #3005 RESULTS
03/07/2026 - 10:04 PM
ACCELERATOR
MODEL
146
tokens/s
generation
112
ms
time to first token
14471
tokens/s
prompt
2663
LocalScore
HOW YOU STACK UP
Explore All ResultsSmollerLM-48M-Instruct-ft-sft - 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
18829
tokens/s
115
tokens/s
63
ms
pp4096+tg256
8847
tokens/s
139
tokens/s
470
ms
pp2048+tg256
14669
tokens/s
147
tokens/s
147
ms
pp2048+tg768
14336
tokens/s
139
tokens/s
149
ms
pp1024+tg1024
22838
tokens/s
140
tokens/s
51
ms
pp1280+tg3072
19620
tokens/s
152
tokens/s
73
ms
pp384+tg1152
23703
tokens/s
163
tokens/s
22
ms
pp64+tg1024
5878
tokens/s
163
tokens/s
18
ms
pp16+tg1536
1521
tokens/s
161
tokens/s
18
ms