TEST #3442 RESULTS
05/04/2026 - 6:31 PM
ACCELERATOR
MODEL
91.0
tokens/s
generation
900
ms
time to first token
1347
tokens/s
prompt
514
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M2 Pro 6P+4E
RAM
16GB
OS
Darwin
Kernel Release
25.4.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.4.0: Thu Mar 19 19:31:17 PDT 2026; root:xnu-12377.101.15~1/RELEASE_ARM64_T6020
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1640
tokens/s
104
tokens/s
634
ms
pp4096+tg256
1390
tokens/s
60.7
tokens/s
2.96
sec
pp2048+tg256
1569
tokens/s
82.7
tokens/s
1.32
sec
pp2048+tg768
1565
tokens/s
79.1
tokens/s
1.32
sec
pp1024+tg1024
1623
tokens/s
93.4
tokens/s
641
ms
pp1280+tg3072
1569
tokens/s
73.9
tokens/s
826
ms
pp384+tg1152
1557
tokens/s
104
tokens/s
255
ms
pp64+tg1024
970
tokens/s
113
tokens/s
74
ms
pp16+tg1536
240
tokens/s
108
tokens/s
74
ms