TEST #863 RESULTS
05/14/2025 - 8:07 AM
ACCELERATOR
MODEL
88.3
tokens/s
generation
955
ms
time to first token
1333
tokens/s
prompt
498
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M3 Pro 5P+6E
RAM
36GB
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 19:54:29 PDT 2025; root:xnu-11417.121.6~2/RELEASE_ARM64_T6030
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1647
tokens/s
97.9
tokens/s
632
ms
pp4096+tg256
1231
tokens/s
62.8
tokens/s
3.34
sec
pp2048+tg256
1491
tokens/s
81.6
tokens/s
1.39
sec
pp2048+tg768
1492
tokens/s
78.7
tokens/s
1.38
sec
pp1024+tg1024
1595
tokens/s
90.3
tokens/s
652
ms
pp1280+tg3072
1579
tokens/s
74.7
tokens/s
821
ms
pp384+tg1152
1655
tokens/s
99.2
tokens/s
241
ms
pp64+tg1024
981
tokens/s
107
tokens/s
73
ms
pp16+tg1536
325
tokens/s
102
tokens/s
58
ms