TEST #3255 RESULTS
04/08/2026 - 4:45 AM
ACCELERATOR
MODEL
48.6
tokens/s
generation
2.09
sec
time to first token
577
tokens/s
prompt
238
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
24.6.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.6.0: Wed Nov 5 21:30:44 PST 2025; root:xnu-11417.140.69.705.2~1/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
778
tokens/s
60.0
tokens/s
1.33
sec
pp4096+tg256
603
tokens/s
40.2
tokens/s
6.82
sec
pp2048+tg256
740
tokens/s
50.1
tokens/s
2.78
sec
pp2048+tg768
616
tokens/s
49.0
tokens/s
3.34
sec
pp1024+tg1024
669
tokens/s
53.7
tokens/s
1.55
sec
pp1280+tg3072
623
tokens/s
46.7
tokens/s
2.07
sec
pp384+tg1152
629
tokens/s
52.7
tokens/s
628
ms
pp64+tg1024
391
tokens/s
46.9
tokens/s
180
ms
pp16+tg1536
142
tokens/s
38.4
tokens/s
130
ms