TEST #1751 RESULTS
11/08/2025 - 12:02 AM
ACCELERATOR
MODEL
75.2
tokens/s
generation
1.20
sec
time to first token
1100
tokens/s
prompt
410
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 4P+6E
RAM
16GB
OS
Darwin
Kernel Release
25.1.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.1.0: Mon Oct 20 19:32:56 PDT 2025; root:xnu-12377.41.6~2/RELEASE_ARM64_T8132
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1300
tokens/s
84.6
tokens/s
799
ms
pp4096+tg256
952
tokens/s
54.5
tokens/s
4.32
sec
pp2048+tg256
1178
tokens/s
69.9
tokens/s
1.75
sec
pp2048+tg768
1183
tokens/s
65.9
tokens/s
1.75
sec
pp1024+tg1024
1305
tokens/s
75.0
tokens/s
796
ms
pp1280+tg3072
1310
tokens/s
64.8
tokens/s
989
ms
pp384+tg1152
1310
tokens/s
84.8
tokens/s
304
ms
pp64+tg1024
1004
tokens/s
90.4
tokens/s
75
ms
pp16+tg1536
360
tokens/s
87.2
tokens/s
56
ms