TEST #805 RESULTS
05/03/2025 - 10:03 PM
ACCELERATOR
MODEL
37.3
tokens/s
generation
2.74
sec
time to first token
445
tokens/s
prompt
182
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M3 Max 10P+4E
RAM
36GB
OS
Darwin
Kernel Release
24.4.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.4.0: Fri Apr 11 18:33:40 PDT 2025; root:xnu-11417.101.15~117/RELEASE_ARM64_T6031
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
530
tokens/s
42.1
tokens/s
1.96
sec
pp4096+tg256
482
tokens/s
30.1
tokens/s
8.52
sec
pp2048+tg256
510
tokens/s
36.5
tokens/s
4.05
sec
pp2048+tg768
511
tokens/s
35.6
tokens/s
4.04
sec
pp1024+tg1024
508
tokens/s
38.8
tokens/s
2.04
sec
pp1280+tg3072
437
tokens/s
34.0
tokens/s
2.96
sec
pp384+tg1152
500
tokens/s
39.9
tokens/s
792
ms
pp64+tg1024
386
tokens/s
39.2
tokens/s
188
ms
pp16+tg1536
145
tokens/s
39.3
tokens/s
132
ms