TEST #309 RESULTS
04/05/2025 - 11:15 PM
ACCELERATOR
MODEL
49.1
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
24.3.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.3.0: Thu Jan 2 20:24:16 PST 2025; root:xnu-11215.81.4~3/RELEASE_ARM64_T6000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
649
tokens/s
52.6
tokens/s
1.60
sec
pp4096+tg256
622
tokens/s
39.5
tokens/s
6.61
sec
pp2048+tg256
643
tokens/s
47.1
tokens/s
3.21
sec
pp2048+tg768
642
tokens/s
46.1
tokens/s
3.21
sec
pp1024+tg1024
641
tokens/s
50.1
tokens/s
1.62
sec
pp1280+tg3072
634
tokens/s
44.4
tokens/s
2.04
sec
pp384+tg1152
578
tokens/s
53.1
tokens/s
682
ms
pp64+tg1024
316
tokens/s
55.2
tokens/s
220
ms
pp16+tg1536
87
tokens/s
54.0
tokens/s
202
ms