TEST #577 RESULTS
04/08/2025 - 10:03 AM
ACCELERATOR
MODEL
89.6
tokens/s
generation
816
ms
time to first token
1556
tokens/s
prompt
555
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M3 Pro 6P+6E
RAM
36GB
OS
Darwin
Kernel Release
24.3.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.3.0: Thu Jan 2 20:24:24 PST 2025; root:xnu-11215.81.4~3/RELEASE_ARM64_T6030
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1995
tokens/s
99.4
tokens/s
523
ms
pp4096+tg256
1406
tokens/s
63.7
tokens/s
2.93
sec
pp2048+tg256
1771
tokens/s
83.0
tokens/s
1.17
sec
pp2048+tg768
1762
tokens/s
79.8
tokens/s
1.17
sec
pp1024+tg1024
1947
tokens/s
91.5
tokens/s
536
ms
pp1280+tg3072
1949
tokens/s
75.8
tokens/s
667
ms
pp384+tg1152
1884
tokens/s
101
tokens/s
213
ms
pp64+tg1024
976
tokens/s
108
tokens/s
74
ms
pp16+tg1536
315
tokens/s
104
tokens/s
59
ms