TEST #34 RESULTS
03/28/2025 - 5:38 PM
ACCELERATOR
MODEL
17.6
tokens/s
generation
6.85
sec
time to first token
188
tokens/s
prompt
78
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 4P+6E
RAM
16GB
OS
Darwin
Kernel Release
24.1.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.1.0: Thu Oct 10 21:06:23 PDT 2024; root:xnu-11215.41.3~3/RELEASE_ARM64_T8132
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
208
tokens/s
18.6
tokens/s
4.97
sec
pp4096+tg256
184
tokens/s
15.0
tokens/s
22.36
sec
pp2048+tg256
200
tokens/s
17.0
tokens/s
10.29
sec
pp2048+tg768
200
tokens/s
16.7
tokens/s
10.29
sec
pp1024+tg1024
208
tokens/s
17.9
tokens/s
4.98
sec
pp1280+tg3072
206
tokens/s
16.4
tokens/s
6.27
sec
pp384+tg1152
211
tokens/s
18.7
tokens/s
1.87
sec
pp64+tg1024
193
tokens/s
19.2
tokens/s
382
ms
pp16+tg1536
82
tokens/s
18.9
tokens/s
246
ms