TEST #441 RESULTS
04/06/2025 - 11:24 PM
ACCELERATOR
MODEL
48.8
tokens/s
generation
2.15
sec
time to first token
535
tokens/s
prompt
230
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M1 Ultra 16P+4E
RAM
128GB
OS
Darwin
Kernel Release
23.6.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 23.6.0: Thu Dec 19 20:44:10 PST 2024; root:xnu-10063.141.1.703.2~1/RELEASE_ARM64_T6000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
646
tokens/s
52.3
tokens/s
1.60
sec
pp4096+tg256
622
tokens/s
39.2
tokens/s
6.61
sec
pp2048+tg256
644
tokens/s
46.8
tokens/s
3.20
sec
pp2048+tg768
642
tokens/s
45.8
tokens/s
3.21
sec
pp1024+tg1024
637
tokens/s
49.9
tokens/s
1.63
sec
pp1280+tg3072
636
tokens/s
44.2
tokens/s
2.03
sec
pp384+tg1152
582
tokens/s
52.8
tokens/s
678
ms
pp64+tg1024
316
tokens/s
55.0
tokens/s
220
ms
pp16+tg1536
89
tokens/s
53.7
tokens/s
197
ms