TEST #1248 RESULTS
07/23/2025 - 8:41 PM
ACCELERATOR
MODEL
62.5
tokens/s
generation
1.73
sec
time to first token
692
tokens/s
prompt
293
LocalScore
HOW YOU STACK UP
Explore All ResultsLLaMA v2 - Q4_0
SYSTEM
CPU
Apple M4 Max 12P+4E
RAM
64GB
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
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
901
tokens/s
68.0
tokens/s
1.15
sec
pp4096+tg256
789
tokens/s
44.8
tokens/s
5.22
sec
pp2048+tg256
767
tokens/s
59.0
tokens/s
2.69
sec
pp2048+tg768
765
tokens/s
56.4
tokens/s
2.69
sec
pp1024+tg1024
790
tokens/s
64.5
tokens/s
1.31
sec
pp1280+tg3072
749
tokens/s
52.8
tokens/s
1.72
sec
pp384+tg1152
741
tokens/s
69.9
tokens/s
532
ms
pp64+tg1024
548
tokens/s
74.9
tokens/s
129
ms
pp16+tg1536
179
tokens/s
72.2
tokens/s
101
ms