TEST #847 RESULTS
05/09/2025 - 9:28 AM
ACCELERATOR
MODEL
56.5
tokens/s
generation
1.67
sec
time to first token
805
tokens/s
prompt
301
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M2 4P+4E
RAM
8GB
OS
Darwin
Kernel Release
24.0.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.0.0: Tue Sep 24 23:37:13 PDT 2024; root:xnu-11215.1.12~1/RELEASE_ARM64_T8112
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
946
tokens/s
62.6
tokens/s
1.10
sec
pp4096+tg256
675
tokens/s
37.4
tokens/s
6.09
sec
pp2048+tg256
829
tokens/s
50.0
tokens/s
2.49
sec
pp2048+tg768
886
tokens/s
47.2
tokens/s
2.33
sec
pp1024+tg1024
970
tokens/s
58.8
tokens/s
1.07
sec
pp1280+tg3072
949
tokens/s
46.3
tokens/s
1.37
sec
pp384+tg1152
990
tokens/s
67.2
tokens/s
400
ms
pp64+tg1024
733
tokens/s
71.0
tokens/s
99
ms
pp16+tg1536
265
tokens/s
67.6
tokens/s
72
ms