TEST #426 RESULTS
04/06/2025 - 9:16 PM
ACCELERATOR
MODEL
34.0
tokens/s
generation
14.46
sec
time to first token
101
tokens/s
prompt
62
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core Ultra 7 258V (arrowlake-s)
RAM
30.3GB
OS
Linux
Kernel Release
6.13.7-061307-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #202503131244 SMP PREEMPT_DYNAMIC Fri Mar 14 02:34:39 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
112
tokens/s
40.2
tokens/s
9.18
sec
pp4096+tg256
84
tokens/s
27.4
tokens/s
48.52
sec
pp2048+tg256
96
tokens/s
32.1
tokens/s
21.43
sec
pp2048+tg768
90
tokens/s
31.5
tokens/s
22.74
sec
pp1024+tg1024
100
tokens/s
33.9
tokens/s
10.24
sec
pp1280+tg3072
95
tokens/s
30.2
tokens/s
13.45
sec
pp384+tg1152
103
tokens/s
36.1
tokens/s
3.76
sec
pp64+tg1024
112
tokens/s
37.6
tokens/s
594
ms
pp16+tg1536
112
tokens/s
36.8
tokens/s
167
ms