TEST #3032 RESULTS
03/07/2026 - 11:13 PM
ACCELERATOR
MODEL
43.5
tokens/s
generation
1.97
sec
time to first token
662
tokens/s
prompt
245
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q5_K - Medium
SYSTEM
CPU
Apple M2 4P+4E
RAM
8GB
OS
Darwin
Kernel Release
24.6.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.6.0: Wed Oct 15 21:12:44 PDT 2025; root:xnu-11417.140.69.703.14~1/RELEASE_ARM64_T8112
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
770
tokens/s
50.0
tokens/s
1.35
sec
pp4096+tg256
608
tokens/s
30.2
tokens/s
6.77
sec
pp2048+tg256
712
tokens/s
39.6
tokens/s
2.90
sec
pp2048+tg768
715
tokens/s
38.9
tokens/s
2.89
sec
pp1024+tg1024
764
tokens/s
44.1
tokens/s
1.36
sec
pp1280+tg3072
756
tokens/s
35.7
tokens/s
1.71
sec
pp384+tg1152
773
tokens/s
49.2
tokens/s
515
ms
pp64+tg1024
619
tokens/s
53.3
tokens/s
119
ms
pp16+tg1536
236
tokens/s
51.0
tokens/s
83
ms