TEST #1566 RESULTS
09/27/2025 - 3:45 PM
ACCELERATOR
MODEL
17.9
tokens/s
generation
8.39
sec
time to first token
149
tokens/s
prompt
68
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
25.0.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.0.0: Mon Aug 25 21:17:53 PDT 2025; root:xnu-12377.1.9~3/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
20.7
tokens/s
6.15
sec
pp4096+tg256
152
tokens/s
14.8
tokens/s
26.95
sec
pp2048+tg256
164
tokens/s
18.9
tokens/s
12.57
sec
pp2048+tg768
163
tokens/s
18.2
tokens/s
12.58
sec
pp1024+tg1024
169
tokens/s
20.2
tokens/s
6.12
sec
pp1280+tg3072
166
tokens/s
17.2
tokens/s
7.75
sec
pp384+tg1152
152
tokens/s
18.2
tokens/s
2.57
sec
pp64+tg1024
149
tokens/s
14.7
tokens/s
476
ms
pp16+tg1536
60
tokens/s
17.9
tokens/s
312
ms