TEST #3995 RESULTS
08/05/2026 - 9:13 AM
ACCELERATOR
MODEL
23.8
tokens/s
generation
5.79
sec
time to first token
216
tokens/s
prompt
96
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M5 Pro 6P+12E
RAM
48GB
OS
Darwin
Kernel Release
25.6.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.6.0: Sat Jul 11 15:27:04 PDT 2026; root:xnu-12377.161.13~4/RELEASE_ARM64_T6050
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
241
tokens/s
25.6
tokens/s
4.30
sec
pp4096+tg256
222
tokens/s
19.8
tokens/s
18.51
sec
pp2048+tg256
235
tokens/s
23.1
tokens/s
8.74
sec
pp2048+tg768
236
tokens/s
22.7
tokens/s
8.71
sec
pp1024+tg1024
242
tokens/s
24.3
tokens/s
4.27
sec
pp1280+tg3072
240
tokens/s
21.9
tokens/s
5.37
sec
pp384+tg1152
242
tokens/s
25.6
tokens/s
1.63
sec
pp64+tg1024
202
tokens/s
26.1
tokens/s
354
ms
pp16+tg1536
80
tokens/s
25.2
tokens/s
236
ms