TEST #452 RESULTS
04/07/2025 - 1:22 AM
ACCELERATOR
MODEL
19.8
tokens/s
generation
7.23
sec
time to first token
170
tokens/s
prompt
78
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M1 Pro 6P+2E
RAM
16GB
OS
Darwin
Kernel Release
24.4.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.4.0: Wed Mar 19 21:16:34 PDT 2025; root:xnu-11417.101.15~1/RELEASE_ARM64_T6000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
196
tokens/s
21.7
tokens/s
5.26
sec
pp4096+tg256
180
tokens/s
15.1
tokens/s
22.79
sec
pp2048+tg256
189
tokens/s
18.3
tokens/s
10.87
sec
pp2048+tg768
189
tokens/s
18.2
tokens/s
10.88
sec
pp1024+tg1024
193
tokens/s
20.1
tokens/s
5.35
sec
pp1280+tg3072
191
tokens/s
17.7
tokens/s
6.75
sec
pp384+tg1152
161
tokens/s
21.7
tokens/s
2.42
sec
pp64+tg1024
167
tokens/s
23.1
tokens/s
423
ms
pp16+tg1536
65
tokens/s
22.4
tokens/s
286
ms