TEST #3445 RESULTS
05/04/2026 - 8:56 PM
ACCELERATOR
MODEL
101
tokens/s
generation
981
ms
time to first token
1266
tokens/s
prompt
507
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M5 4P+6E
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 19:33:43 PDT 2026; root:xnu-12377.101.15~1/RELEASE_ARM64_T8142
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1566
tokens/s
111
tokens/s
663
ms
pp4096+tg256
1235
tokens/s
72.7
tokens/s
3.33
sec
pp2048+tg256
1430
tokens/s
93.9
tokens/s
1.44
sec
pp2048+tg768
1431
tokens/s
90.7
tokens/s
1.44
sec
pp1024+tg1024
1517
tokens/s
103
tokens/s
684
ms
pp1280+tg3072
1503
tokens/s
85.8
tokens/s
861
ms
pp384+tg1152
1496
tokens/s
113
tokens/s
265
ms
pp64+tg1024
916
tokens/s
120
tokens/s
78
ms
pp16+tg1536
301
tokens/s
116
tokens/s
61
ms