TEST #1105 RESULTS
06/24/2025 - 11:04 PM
ACCELERATOR
MODEL
7.9
tokens/s
generation
10.84
sec
time to first token
113
tokens/s
prompt
44
LocalScore
HOW YOU STACK UP
Explore All ResultsModels Mlx Community Meta Llama 3.1 8B Instruct Bf16 - Q6_K
SYSTEM
CPU
Apple M3 4P+4E
RAM
24GB
OS
Darwin
Kernel Release
24.5.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.5.0: Tue Apr 22 19:54:33 PDT 2025; root:xnu-11417.121.6~2/RELEASE_ARM64_T8122
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
159
tokens/s
12.7
tokens/s
6.52
sec
pp4096+tg256
126
tokens/s
10.2
tokens/s
32.49
sec
pp2048+tg256
124
tokens/s
10.4
tokens/s
16.55
sec
pp2048+tg768
124
tokens/s
6.4
tokens/s
16.67
sec
pp1024+tg1024
121
tokens/s
6.5
tokens/s
8.59
sec
pp1280+tg3072
110
tokens/s
6.2
tokens/s
11.69
sec
pp384+tg1152
104
tokens/s
6.4
tokens/s
3.82
sec
pp64+tg1024
97
tokens/s
6.7
tokens/s
787
ms
pp16+tg1536
53
tokens/s
5.9
tokens/s
420
ms