TEST #1707 RESULTS
10/27/2025 - 12:30 PM
ACCELERATOR
MODEL
24.3
tokens/s
generation
5.56
sec
time to first token
226
tokens/s
prompt
100
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M2 Pro 6P+4E
RAM
16GB
OS
Darwin
Kernel Release
24.6.0
Architecture
arm64
Version
Cosmopolitan 3.9.7 MODE=aarch64; Darwin Kernel Version 24.6.0: Mon Jul 14 11:30:30 PDT 2025; root:xnu-11417.140.69~1/RELEASE_ARM64_T6020
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
249
tokens/s
24.8
tokens/s
4.14
sec
pp4096+tg256
230
tokens/s
18.7
tokens/s
17.82
sec
pp2048+tg256
241
tokens/s
23.3
tokens/s
8.54
sec
pp2048+tg768
251
tokens/s
22.5
tokens/s
8.20
sec
pp1024+tg1024
252
tokens/s
25.2
tokens/s
4.10
sec
pp1280+tg3072
251
tokens/s
22.0
tokens/s
5.13
sec
pp384+tg1152
251
tokens/s
26.9
tokens/s
1.56
sec
pp64+tg1024
214
tokens/s
27.8
tokens/s
333
ms
pp16+tg1536
89
tokens/s
27.7
tokens/s
213
ms