TEST #3129 RESULTS
03/20/2026 - 1:06 PM
ACCELERATOR
MODEL
15.8
tokens/s
generation
11.03
sec
time to first token
116
tokens/s
prompt
55
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple A18 Pro 2P+4E
RAM
8GB
OS
Darwin
Kernel Release
25.3.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.3.0: Wed Jan 28 20:56:11 PST 2026; root:xnu-12377.91.3~2/RELEASE_ARM64_T8140
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
139
tokens/s
19.4
tokens/s
7.43
sec
pp4096+tg256
124
tokens/s
14.3
tokens/s
33.17
sec
pp2048+tg256
121
tokens/s
16.2
tokens/s
16.95
sec
pp2048+tg768
122
tokens/s
15.2
tokens/s
16.80
sec
pp1024+tg1024
116
tokens/s
16.1
tokens/s
8.91
sec
pp1280+tg3072
111
tokens/s
13.2
tokens/s
11.55
sec
pp384+tg1152
109
tokens/s
15.9
tokens/s
3.60
sec
pp64+tg1024
132
tokens/s
16.9
tokens/s
541
ms
pp16+tg1536
65
tokens/s
15.3
tokens/s
296
ms