TEST #1938 RESULTS
12/13/2025 - 10:29 PM
ACCELERATOR
MODEL
120
tokens/s
generation
312
ms
time to first token
4474
tokens/s
prompt
1197
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E3-1245 v5 @ 3.50GHz (skylake)
RAM
47.8GB
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
5475
tokens/s
124
tokens/s
196
ms
pp4096+tg256
3631
tokens/s
103
tokens/s
1.14
sec
pp2048+tg256
4647
tokens/s
117
tokens/s
450
ms
pp2048+tg768
4639
tokens/s
114
tokens/s
450
ms
pp1024+tg1024
5423
tokens/s
122
tokens/s
197
ms
pp1280+tg3072
5101
tokens/s
111
tokens/s
260
ms
pp384+tg1152
6176
tokens/s
127
tokens/s
71
ms
pp64+tg1024
4369
tokens/s
130
tokens/s
21
ms
pp16+tg1536
802
tokens/s
129
tokens/s
27
ms