TEST #1855 RESULTS
11/23/2025 - 2:11 PM
ACCELERATOR
MODEL
14.0
tokens/s
generation
9.49
sec
time to first token
135
tokens/s
prompt
58
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M2 4P+4E
RAM
16GB
OS
Darwin
Kernel Release
25.0.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.0.0: Wed Sep 17 21:38:03 PDT 2025; root:xnu-12377.1.9~141/RELEASE_ARM64_T8112
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
155
tokens/s
15.3
tokens/s
6.67
sec
pp4096+tg256
134
tokens/s
11.4
tokens/s
30.71
sec
pp2048+tg256
146
tokens/s
13.7
tokens/s
14.14
sec
pp2048+tg768
146
tokens/s
13.0
tokens/s
14.13
sec
pp1024+tg1024
142
tokens/s
13.5
tokens/s
7.30
sec
pp1280+tg3072
143
tokens/s
12.8
tokens/s
9.01
sec
pp384+tg1152
152
tokens/s
15.2
tokens/s
2.59
sec
pp64+tg1024
139
tokens/s
15.6
tokens/s
518
ms
pp16+tg1536
60
tokens/s
15.4
tokens/s
326
ms