TEST #393 RESULTS
04/06/2025 - 4:53 PM
ACCELERATOR
MODEL
43.0
tokens/s
generation
2.15
sec
time to first token
526
tokens/s
prompt
219
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Max 12P+4E
RAM
128GB
OS
Darwin
Kernel Release
24.4.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.4.0: Wed Mar 19 21:17:35 PDT 2025; root:xnu-11417.101.15~1/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
810
tokens/s
62.3
tokens/s
1.28
sec
pp4096+tg256
700
tokens/s
43.5
tokens/s
5.88
sec
pp2048+tg256
655
tokens/s
52.9
tokens/s
3.15
sec
pp2048+tg768
589
tokens/s
45.9
tokens/s
3.50
sec
pp1024+tg1024
633
tokens/s
44.9
tokens/s
1.63
sec
pp1280+tg3072
515
tokens/s
20.8
tokens/s
2.51
sec
pp384+tg1152
363
tokens/s
35.6
tokens/s
1.09
sec
pp64+tg1024
342
tokens/s
39.8
tokens/s
207
ms
pp16+tg1536
130
tokens/s
41.5
tokens/s
142
ms