TEST #4115 RESULTS
09/02/2026 - 5:02 PM
ACCELERATOR
MODEL
17.7
tokens/s
generation
6.95
sec
time to first token
184
tokens/s
prompt
78
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 4P+6E
RAM
16GB
OS
Darwin
Kernel Release
27.0.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 27.0.0: Tue Aug 11 22:02:53 PDT 2026; root:xnu-13432.1.9~3/RELEASE_ARM64_T8132
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
208
tokens/s
18.8
tokens/s
4.97
sec
pp4096+tg256
184
tokens/s
14.9
tokens/s
22.36
sec
pp2048+tg256
200
tokens/s
17.2
tokens/s
10.27
sec
pp2048+tg768
198
tokens/s
16.9
tokens/s
10.43
sec
pp1024+tg1024
198
tokens/s
18.0
tokens/s
5.22
sec
pp1280+tg3072
193
tokens/s
16.5
tokens/s
6.67
sec
pp384+tg1152
199
tokens/s
18.8
tokens/s
1.98
sec
pp64+tg1024
191
tokens/s
19.3
tokens/s
390
ms
pp16+tg1536
84
tokens/s
18.9
tokens/s
241
ms