TEST #1930 RESULTS
12/12/2025 - 6:04 AM
ACCELERATOR
MODEL
32.5
tokens/s
generation
3.49
sec
time to first token
357
tokens/s
prompt
149
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Pro 10P+4E
RAM
48GB
OS
Darwin
Kernel Release
25.1.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 25.1.0: Mon Oct 20 19:34:05 PDT 2025; root:xnu-12377.41.6~2/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
405
tokens/s
35.1
tokens/s
2.56
sec
pp4096+tg256
369
tokens/s
25.0
tokens/s
11.13
sec
pp2048+tg256
395
tokens/s
30.8
tokens/s
5.22
sec
pp2048+tg768
394
tokens/s
30.0
tokens/s
5.23
sec
pp1024+tg1024
393
tokens/s
33.2
tokens/s
2.63
sec
pp1280+tg3072
394
tokens/s
28.9
tokens/s
3.28
sec
pp384+tg1152
409
tokens/s
35.9
tokens/s
965
ms
pp64+tg1024
343
tokens/s
37.6
tokens/s
211
ms
pp16+tg1536
113
tokens/s
36.0
tokens/s
166
ms