TEST #1184 RESULTS
07/07/2025 - 1:56 AM
ACCELERATOR
MODEL
104
tokens/s
generation
793
ms
time to first token
1508
tokens/s
prompt
582
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M2 Pro 8P+4E
RAM
32GB
OS
Darwin
Kernel Release
24.5.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.5.0: Tue Apr 22 19:54:25 PDT 2025; root:xnu-11417.121.6~2/RELEASE_ARM64_T6020
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1859
tokens/s
117
tokens/s
559
ms
pp4096+tg256
1572
tokens/s
69.1
tokens/s
2.62
sec
pp2048+tg256
1794
tokens/s
94.0
tokens/s
1.15
sec
pp2048+tg768
1792
tokens/s
90.0
tokens/s
1.15
sec
pp1024+tg1024
1843
tokens/s
106
tokens/s
564
ms
pp1280+tg3072
1812
tokens/s
84.5
tokens/s
715
ms
pp384+tg1152
1716
tokens/s
119
tokens/s
231
ms
pp64+tg1024
892
tokens/s
131
tokens/s
79
ms
pp16+tg1536
287
tokens/s
124
tokens/s
63
ms