TEST #1102 RESULTS
06/24/2025 - 1:42 PM
ACCELERATOR
MODEL
78.7
tokens/s
generation
1.03
sec
time to first token
1150
tokens/s
prompt
445
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M1 Pro 8P+2E
RAM
16GB
OS
Darwin
Kernel Release
24.5.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.5.0: Tue Apr 22 19:54:49 PDT 2025; root:xnu-11417.121.6~2/RELEASE_ARM64_T6000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1452
tokens/s
89.1
tokens/s
716
ms
pp4096+tg256
1231
tokens/s
52.4
tokens/s
3.35
sec
pp2048+tg256
1370
tokens/s
71.3
tokens/s
1.51
sec
pp2048+tg768
1352
tokens/s
68.3
tokens/s
1.53
sec
pp1024+tg1024
1437
tokens/s
80.4
tokens/s
724
ms
pp1280+tg3072
1377
tokens/s
64.4
tokens/s
941
ms
pp384+tg1152
1274
tokens/s
90.7
tokens/s
311
ms
pp64+tg1024
614
tokens/s
99.1
tokens/s
113
ms
pp16+tg1536
239
tokens/s
92.9
tokens/s
76
ms