TEST #3342 RESULTS
04/18/2026 - 5:36 AM
ACCELERATOR
MODEL
27.5
tokens/s
generation
15.40
sec
time to first token
93
tokens/s
prompt
55
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
12th Gen Intel Core i7-1265U (alderlake)
RAM
30.8GB
OS
Linux
Kernel Release
6.17.0-22-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #22-Ubuntu SMP PREEMPT_DYNAMIC Fri Mar 13 12:04:44 UTC 2026
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
104
tokens/s
33.9
tokens/s
9.86
sec
pp4096+tg256
78
tokens/s
20.6
tokens/s
52.87
sec
pp2048+tg256
87
tokens/s
27.3
tokens/s
23.61
sec
pp2048+tg768
89
tokens/s
25.8
tokens/s
23.02
sec
pp1024+tg1024
97
tokens/s
28.6
tokens/s
10.62
sec
pp1280+tg3072
94
tokens/s
23.6
tokens/s
13.66
sec
pp384+tg1152
95
tokens/s
28.7
tokens/s
4.07
sec
pp64+tg1024
103
tokens/s
29.6
tokens/s
655
ms
pp16+tg1536
94
tokens/s
29.0
tokens/s
200
ms