TEST #2018 RESULTS
12/21/2025 - 10:04 PM
ACCELERATOR
MODEL
28.5
tokens/s
generation
4.02
sec
time to first token
309
tokens/s
prompt
130
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Max 12P+4E
RAM
64GB
OS
Darwin
Kernel Release
25.1.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.1.0: Mon Oct 20 19:34:05 PDT 2025; root:xnu-12377.41.6~2/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
396
tokens/s
34.2
tokens/s
2.62
sec
pp4096+tg256
319
tokens/s
23.0
tokens/s
12.88
sec
pp2048+tg256
343
tokens/s
27.7
tokens/s
6.02
sec
pp2048+tg768
332
tokens/s
27.3
tokens/s
6.20
sec
pp1024+tg1024
342
tokens/s
30.3
tokens/s
3.03
sec
pp1280+tg3072
337
tokens/s
26.3
tokens/s
3.83
sec
pp384+tg1152
332
tokens/s
29.5
tokens/s
1.19
sec
pp64+tg1024
286
tokens/s
29.7
tokens/s
250
ms
pp16+tg1536
97
tokens/s
28.4
tokens/s
191
ms