TEST #470 RESULTS
04/07/2025 - 4:26 AM
ACCELERATOR
MODEL
98.0
tokens/s
generation
819
ms
time to first token
1477
tokens/s
prompt
561
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M2 Pro 8P+4E
RAM
32GB
OS
Darwin
Kernel Release
23.2.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 23.2.0: Wed Nov 15 21:55:06 PST 2023; root:xnu-10002.61.3~2/RELEASE_ARM64_T6020
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1812
tokens/s
113
tokens/s
574
ms
pp4096+tg256
1507
tokens/s
64.7
tokens/s
2.73
sec
pp2048+tg256
1761
tokens/s
89.2
tokens/s
1.17
sec
pp2048+tg768
1744
tokens/s
84.8
tokens/s
1.19
sec
pp1024+tg1024
1801
tokens/s
101
tokens/s
578
ms
pp1280+tg3072
1711
tokens/s
79.6
tokens/s
757
ms
pp384+tg1152
1723
tokens/s
111
tokens/s
231
ms
pp64+tg1024
941
tokens/s
123
tokens/s
75
ms
pp16+tg1536
290
tokens/s
116
tokens/s
63
ms