TEST #1522 RESULTS
09/15/2025 - 9:42 PM
ACCELERATOR
MODEL
56.1
tokens/s
generation
1.72
sec
time to first token
704
tokens/s
prompt
284
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Max 12P+4E
RAM
128GB
OS
Darwin
Kernel Release
25.0.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.0.0: Mon Aug 25 21:17:54 PDT 2025; root:xnu-12377.1.9~3/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
821
tokens/s
62.1
tokens/s
1.26
sec
pp4096+tg256
754
tokens/s
42.9
tokens/s
5.46
sec
pp2048+tg256
798
tokens/s
53.0
tokens/s
2.58
sec
pp2048+tg768
798
tokens/s
51.6
tokens/s
2.58
sec
pp1024+tg1024
814
tokens/s
57.4
tokens/s
1.27
sec
pp1280+tg3072
807
tokens/s
49.4
tokens/s
1.60
sec
pp384+tg1152
790
tokens/s
61.4
tokens/s
501
ms
pp64+tg1024
547
tokens/s
64.5
tokens/s
132
ms
pp16+tg1536
202
tokens/s
62.8
tokens/s
94
ms