TEST #1396 RESULTS
08/19/2025 - 8:44 AM
ACCELERATOR
MODEL
25.4
tokens/s
generation
24.71
sec
time to first token
60
tokens/s
prompt
39
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
12th Gen Intel Core i9-12900HX (alderlake)
RAM
127.8GB
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
60
tokens/s
27.4
tokens/s
17.20
sec
pp4096+tg256
48
tokens/s
20.2
tokens/s
84.82
sec
pp2048+tg256
56
tokens/s
24.4
tokens/s
36.78
sec
pp2048+tg768
55
tokens/s
23.5
tokens/s
37.21
sec
pp1024+tg1024
59
tokens/s
25.9
tokens/s
17.25
sec
pp1280+tg3072
59
tokens/s
22.9
tokens/s
21.86
sec
pp384+tg1152
64
tokens/s
27.5
tokens/s
6.05
sec
pp64+tg1024
74
tokens/s
28.7
tokens/s
901
ms
pp16+tg1536
63
tokens/s
28.1
tokens/s
288
ms