TEST #1747 RESULTS
11/07/2025 - 10:31 PM
ACCELERATOR
MODEL
85.6
tokens/s
generation
826
ms
time to first token
1574
tokens/s
prompt
546
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M3 Pro 6P+6E
RAM
18GB
OS
Darwin
Kernel Release
24.6.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.6.0: Mon Aug 11 21:16:31 PDT 2025; root:xnu-11417.140.69.701.11~1/RELEASE_ARM64_T6030
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1932
tokens/s
90.9
tokens/s
542
ms
pp4096+tg256
1374
tokens/s
62.7
tokens/s
3.00
sec
pp2048+tg256
1768
tokens/s
82.2
tokens/s
1.17
sec
pp2048+tg768
1775
tokens/s
77.0
tokens/s
1.17
sec
pp1024+tg1024
1951
tokens/s
87.0
tokens/s
535
ms
pp1280+tg3072
1843
tokens/s
75.1
tokens/s
705
ms
pp384+tg1152
1967
tokens/s
98.0
tokens/s
204
ms
pp64+tg1024
1135
tokens/s
103
tokens/s
67
ms
pp16+tg1536
419
tokens/s
94.9
tokens/s
48
ms