TEST #563 RESULTS
04/08/2025 - 6:59 AM
ACCELERATOR
MODEL
31.6
tokens/s
generation
3.45
sec
time to first token
351
tokens/s
prompt
148
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Pro 10P+4E
RAM
24GB
OS
Darwin
Kernel Release
24.2.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.2.0: Fri Dec 6 19:03:40 PST 2024; root:xnu-11215.61.5~2/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
412
tokens/s
36.0
tokens/s
2.51
sec
pp4096+tg256
380
tokens/s
25.3
tokens/s
10.83
sec
pp2048+tg256
403
tokens/s
31.3
tokens/s
5.12
sec
pp2048+tg768
405
tokens/s
30.3
tokens/s
5.09
sec
pp1024+tg1024
404
tokens/s
33.7
tokens/s
2.56
sec
pp1280+tg3072
394
tokens/s
25.3
tokens/s
3.28
sec
pp384+tg1152
319
tokens/s
31.7
tokens/s
1.24
sec
pp64+tg1024
320
tokens/s
35.5
tokens/s
226
ms
pp16+tg1536
124
tokens/s
35.4
tokens/s
154
ms