TEST #1560 RESULTS
09/26/2025 - 11:30 AM
ACCELERATOR
MODEL
67.0
tokens/s
generation
1.42
sec
time to first token
927
tokens/s
prompt
353
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M3 4P+4E
RAM
24GB
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 21:16:33 PDT 2025; root:xnu-11417.140.69.701.11~1/RELEASE_ARM64_T8122
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1117
tokens/s
74.2
tokens/s
930
ms
pp4096+tg256
812
tokens/s
47.9
tokens/s
5.07
sec
pp2048+tg256
1005
tokens/s
62.2
tokens/s
2.05
sec
pp2048+tg768
1003
tokens/s
59.9
tokens/s
2.06
sec
pp1024+tg1024
1116
tokens/s
68.6
tokens/s
931
ms
pp1280+tg3072
1092
tokens/s
56.8
tokens/s
1.19
sec
pp384+tg1152
1069
tokens/s
75.2
tokens/s
371
ms
pp64+tg1024
826
tokens/s
80.6
tokens/s
89
ms
pp16+tg1536
303
tokens/s
77.3
tokens/s
64
ms