TEST #1140 RESULTS
06/28/2025 - 6:02 AM
ACCELERATOR
MODEL
18.8
tokens/s
generation
8.40
sec
time to first token
148
tokens/s
prompt
69
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M2 4P+4E
RAM
8GB
OS
Darwin
Kernel Release
24.5.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.5.0: Tue Apr 22 19:54:26 PDT 2025; root:xnu-11417.121.6~2/RELEASE_ARM64_T8112
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
169
tokens/s
21.6
tokens/s
6.13
sec
pp4096+tg256
153
tokens/s
15.2
tokens/s
26.88
sec
pp2048+tg256
163
tokens/s
19.1
tokens/s
12.59
sec
pp2048+tg768
164
tokens/s
18.4
tokens/s
12.57
sec
pp1024+tg1024
169
tokens/s
20.6
tokens/s
6.12
sec
pp1280+tg3072
166
tokens/s
16.5
tokens/s
7.74
sec
pp384+tg1152
143
tokens/s
18.1
tokens/s
2.74
sec
pp64+tg1024
141
tokens/s
20.6
tokens/s
499
ms
pp16+tg1536
68
tokens/s
19.2
tokens/s
277
ms