TEST #3359 RESULTS
04/21/2026 - 4:18 PM
ACCELERATOR
MODEL
17.8
tokens/s
generation
7.00
sec
time to first token
182
tokens/s
prompt
77
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
Apple M4 4P+6E
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 Jan 19 22:01:41 PST 2026; root:xnu-11417.140.69.708.3~1/RELEASE_ARM64_T8132
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
209
tokens/s
18.8
tokens/s
4.96
sec
pp4096+tg256
184
tokens/s
15.1
tokens/s
22.32
sec
pp2048+tg256
198
tokens/s
17.4
tokens/s
10.39
sec
pp2048+tg768
193
tokens/s
17.0
tokens/s
10.70
sec
pp1024+tg1024
195
tokens/s
18.1
tokens/s
5.29
sec
pp1280+tg3072
193
tokens/s
16.5
tokens/s
6.68
sec
pp384+tg1152
199
tokens/s
18.9
tokens/s
1.98
sec
pp64+tg1024
185
tokens/s
19.4
tokens/s
395
ms
pp16+tg1536
83
tokens/s
19.1
tokens/s
243
ms