TEST #3897 RESULTS
07/13/2026 - 1:51 AM
ACCELERATOR
MODEL
14.5
tokens/s
generation
8.98
sec
time to first token
143
tokens/s
prompt
61
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M2 4P+4E
RAM
16GB
OS
Darwin
Kernel Release
25.4.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.4.0: Thu Mar 19 19:33:09 PDT 2026; root:xnu-12377.101.15~1/RELEASE_ARM64_T8112
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
159
tokens/s
15.4
tokens/s
6.52
sec
pp4096+tg256
140
tokens/s
11.7
tokens/s
29.35
sec
pp2048+tg256
153
tokens/s
13.8
tokens/s
13.48
sec
pp2048+tg768
153
tokens/s
13.8
tokens/s
13.49
sec
pp1024+tg1024
159
tokens/s
15.2
tokens/s
6.51
sec
pp1280+tg3072
157
tokens/s
13.0
tokens/s
8.21
sec
pp384+tg1152
159
tokens/s
15.5
tokens/s
2.47
sec
pp64+tg1024
144
tokens/s
16.2
tokens/s
507
ms
pp16+tg1536
63
tokens/s
16.3
tokens/s
312
ms