TEST #780 RESULTS
04/29/2025 - 11:40 AM
ACCELERATOR
MODEL
24.5
tokens/s
generation
14.85
sec
time to first token
95
tokens/s
prompt
54
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Core i7-7700HQ CPU @ 2.80GHz (skylake)
RAM
15.5GB
OS
Linux
Kernel Release
6.14.4-arch1-1
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Sat, 26 Apr 2025 00:06:37 +0000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
121
tokens/s
27.9
tokens/s
8.48
sec
pp4096+tg256
88
tokens/s
21.0
tokens/s
46.40
sec
pp2048+tg256
102
tokens/s
24.4
tokens/s
20.16
sec
pp2048+tg768
95
tokens/s
21.5
tokens/s
21.60
sec
pp1024+tg1024
53
tokens/s
21.4
tokens/s
19.46
sec
pp1280+tg3072
99
tokens/s
21.4
tokens/s
12.92
sec
pp384+tg1152
101
tokens/s
26.4
tokens/s
3.82
sec
pp64+tg1024
111
tokens/s
28.2
tokens/s
612
ms
pp16+tg1536
88
tokens/s
27.9
tokens/s
216
ms