TEST #3285 RESULTS
04/11/2026 - 9:17 AM
ACCELERATOR
MODEL
47.7
tokens/s
generation
2.18
sec
time to first token
562
tokens/s
prompt
231
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Max 10P+4E
RAM
36GB
OS
Darwin
Kernel Release
25.4.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.4.0: Thu Mar 19 19:33:25 PDT 2026; root:xnu-12377.101.15~1/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
653
tokens/s
51.4
tokens/s
1.59
sec
pp4096+tg256
596
tokens/s
37.9
tokens/s
6.89
sec
pp2048+tg256
636
tokens/s
45.8
tokens/s
3.24
sec
pp2048+tg768
637
tokens/s
44.7
tokens/s
3.24
sec
pp1024+tg1024
632
tokens/s
49.3
tokens/s
1.64
sec
pp1280+tg3072
597
tokens/s
43.1
tokens/s
2.16
sec
pp384+tg1152
636
tokens/s
52.2
tokens/s
621
ms
pp64+tg1024
489
tokens/s
53.8
tokens/s
147
ms
pp16+tg1536
180
tokens/s
51.2
tokens/s
107
ms