TEST #1655 RESULTS
10/14/2025 - 8:47 PM
ACCELERATOR
MODEL
56.2
tokens/s
generation
11.45
sec
time to first token
174
tokens/s
prompt
95
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 5900X 12-Core Processor (znver3)
RAM
63.9GB
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
120
tokens/s
68.9
tokens/s
8.57
sec
pp4096+tg256
113
tokens/s
25.1
tokens/s
36.40
sec
pp2048+tg256
117
tokens/s
47.2
tokens/s
17.48
sec
pp2048+tg768
117
tokens/s
43.8
tokens/s
17.46
sec
pp1024+tg1024
120
tokens/s
55.9
tokens/s
8.58
sec
pp1280+tg3072
119
tokens/s
37.9
tokens/s
10.76
sec
pp384+tg1152
121
tokens/s
69.5
tokens/s
3.20
sec
pp64+tg1024
123
tokens/s
81.7
tokens/s
534
ms
pp16+tg1536
614
tokens/s
76.2
tokens/s
38
ms