TEST #1249 RESULTS
07/24/2025 - 2:31 PM
ACCELERATOR
MODEL
13.3
tokens/s
generation
11.05
sec
time to first token
118
tokens/s
prompt
52
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Phenom(tm) II X4 945 Processor (amdfam10)
RAM
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
124
tokens/s
13.6
tokens/s
8.30
sec
pp4096+tg256
117
tokens/s
9.5
tokens/s
35.16
sec
pp2048+tg256
123
tokens/s
12.4
tokens/s
16.69
sec
pp2048+tg768
123
tokens/s
12.0
tokens/s
16.72
sec
pp1024+tg1024
125
tokens/s
13.7
tokens/s
8.23
sec
pp1280+tg3072
124
tokens/s
11.3
tokens/s
10.40
sec
pp384+tg1152
126
tokens/s
15.1
tokens/s
3.11
sec
pp64+tg1024
111
tokens/s
16.2
tokens/s
631
ms
pp16+tg1536
88
tokens/s
15.6
tokens/s
237
ms