TEST #382 RESULTS
04/06/2025 - 3:25 PM
ACCELERATOR
MODEL
30.3
tokens/s
generation
4.27
sec
time to first token
292
tokens/s
prompt
128
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 Pro 8P+4E
RAM
24GB
OS
Darwin
Kernel Release
24.4.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.4.0: Wed Mar 19 21:17:35 PDT 2025; root:xnu-11417.101.15~1/RELEASE_ARM64_T6041
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
336
tokens/s
32.8
tokens/s
3.08
sec
pp4096+tg256
309
tokens/s
23.6
tokens/s
13.28
sec
pp2048+tg256
327
tokens/s
28.9
tokens/s
6.29
sec
pp2048+tg768
317
tokens/s
28.0
tokens/s
6.49
sec
pp1024+tg1024
304
tokens/s
30.7
tokens/s
3.40
sec
pp1280+tg3072
303
tokens/s
26.9
tokens/s
4.26
sec
pp384+tg1152
316
tokens/s
33.3
tokens/s
1.24
sec
pp64+tg1024
296
tokens/s
34.9
tokens/s
244
ms
pp16+tg1536
123
tokens/s
34.1
tokens/s
157
ms