TEST #1569 RESULTS
09/29/2025 - 6:39 PM
ACCELERATOR
MODEL
105
tokens/s
generation
688
ms
time to first token
1680
tokens/s
prompt
635
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M1 Max 8P+2E
RAM
64GB
OS
Darwin
Kernel Release
24.6.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.6.0: Mon Jul 14 11:30:29 PDT 2025; root:xnu-11417.140.69~1/RELEASE_ARM64_T6000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2100
tokens/s
119
tokens/s
496
ms
pp4096+tg256
1908
tokens/s
73.1
tokens/s
2.16
sec
pp2048+tg256
2061
tokens/s
97.3
tokens/s
1.00
sec
pp2048+tg768
2062
tokens/s
92.6
tokens/s
1.00
sec
pp1024+tg1024
2001
tokens/s
108
tokens/s
520
ms
pp1280+tg3072
2058
tokens/s
88.3
tokens/s
631
ms
pp384+tg1152
1810
tokens/s
121
tokens/s
219
ms
pp64+tg1024
907
tokens/s
127
tokens/s
78
ms
pp16+tg1536
217
tokens/s
119
tokens/s
81
ms