TEST #3123 RESULTS
03/19/2026 - 9:17 PM
ACCELERATOR
MODEL
60.3
tokens/s
generation
1.62
sec
time to first token
804
tokens/s
prompt
310
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M3 4P+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 5 22:46:43 PST 2026; root:xnu-12377.101.14~19/RELEASE_ARM64_T8122
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
943
tokens/s
67.4
tokens/s
1.10
sec
pp4096+tg256
730
tokens/s
42.1
tokens/s
5.64
sec
pp2048+tg256
872
tokens/s
55.6
tokens/s
2.37
sec
pp2048+tg768
868
tokens/s
53.2
tokens/s
2.38
sec
pp1024+tg1024
936
tokens/s
61.6
tokens/s
1.11
sec
pp1280+tg3072
922
tokens/s
50.5
tokens/s
1.40
sec
pp384+tg1152
887
tokens/s
68.3
tokens/s
446
ms
pp64+tg1024
777
tokens/s
73.6
tokens/s
95
ms
pp16+tg1536
299
tokens/s
70.6
tokens/s
66
ms