TEST #1666 RESULTS
10/18/2025 - 1:32 AM
ACCELERATOR
MODEL
52.4
tokens/s
generation
1.85
sec
time to first token
636
tokens/s
prompt
262
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Max 12P+4E
RAM
64GB
OS
Darwin
Kernel Release
24.6.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.6.0: Mon Jul 14 11:30:40 PDT 2025; root:xnu-11417.140.69~1/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
817
tokens/s
60.0
tokens/s
1.27
sec
pp4096+tg256
745
tokens/s
42.5
tokens/s
5.52
sec
pp2048+tg256
784
tokens/s
52.4
tokens/s
2.63
sec
pp2048+tg768
719
tokens/s
51.8
tokens/s
2.87
sec
pp1024+tg1024
644
tokens/s
56.6
tokens/s
1.61
sec
pp1280+tg3072
655
tokens/s
48.7
tokens/s
1.97
sec
pp384+tg1152
730
tokens/s
55.2
tokens/s
542
ms
pp64+tg1024
478
tokens/s
54.8
tokens/s
149
ms
pp16+tg1536
156
tokens/s
49.7
tokens/s
118
ms