TEST #3376 RESULTS
04/24/2026 - 12:26 PM
ACCELERATOR
MODEL
110
tokens/s
generation
678
ms
time to first token
1780
tokens/s
prompt
662
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Pro 8P+4E
RAM
24GB
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 19:33:25 PDT 2026; root:xnu-12377.101.15~1/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2202
tokens/s
124
tokens/s
473
ms
pp4096+tg256
1850
tokens/s
73.7
tokens/s
2.23
sec
pp2048+tg256
2095
tokens/s
101
tokens/s
987
ms
pp2048+tg768
2063
tokens/s
96.0
tokens/s
1.00
sec
pp1024+tg1024
2124
tokens/s
112
tokens/s
490
ms
pp1280+tg3072
2118
tokens/s
90.1
tokens/s
613
ms
pp384+tg1152
2017
tokens/s
127
tokens/s
197
ms
pp64+tg1024
1206
tokens/s
138
tokens/s
60
ms
pp16+tg1536
343
tokens/s
131
tokens/s
53
ms