TEST #1253 RESULTS
07/25/2025 - 4:12 AM
ACCELERATOR
MODEL
55.5
tokens/s
generation
1.81
sec
time to first token
655
tokens/s
prompt
272
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Max 12P+4E
RAM
48GB
OS
Darwin
Kernel Release
24.5.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.5.0: Tue Apr 22 19:53:27 PDT 2025; root:xnu-11417.121.6~2/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
816
tokens/s
60.3
tokens/s
1.27
sec
pp4096+tg256
752
tokens/s
42.8
tokens/s
5.47
sec
pp2048+tg256
792
tokens/s
52.7
tokens/s
2.60
sec
pp2048+tg768
758
tokens/s
51.3
tokens/s
2.72
sec
pp1024+tg1024
690
tokens/s
56.5
tokens/s
1.50
sec
pp1280+tg3072
678
tokens/s
49.0
tokens/s
1.91
sec
pp384+tg1152
750
tokens/s
60.7
tokens/s
527
ms
pp64+tg1024
496
tokens/s
63.9
tokens/s
144
ms
pp16+tg1536
167
tokens/s
62.2
tokens/s
110
ms