TEST #2019 RESULTS
12/21/2025 - 11:50 PM
ACCELERATOR
MODEL
29.1
tokens/s
generation
3.93
sec
time to first token
321
tokens/s
prompt
133
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
400
tokens/s
33.3
tokens/s
2.59
sec
pp4096+tg256
325
tokens/s
22.5
tokens/s
12.65
sec
pp2048+tg256
353
tokens/s
27.4
tokens/s
5.84
sec
pp2048+tg768
342
tokens/s
26.7
tokens/s
6.02
sec
pp1024+tg1024
348
tokens/s
29.4
tokens/s
2.98
sec
pp1280+tg3072
343
tokens/s
26.1
tokens/s
3.77
sec
pp384+tg1152
359
tokens/s
31.4
tokens/s
1.10
sec
pp64+tg1024
310
tokens/s
32.8
tokens/s
233
ms
pp16+tg1536
107
tokens/s
32.1
tokens/s
176
ms