TEST #1772 RESULTS
11/09/2025 - 10:36 PM
ACCELERATOR
MODEL
33.2
tokens/s
generation
679
ms
time to first token
2035
tokens/s
prompt
463
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
RAM
119.7GB
OS
Linux
Kernel Release
6.11.0-1016-nvidia
Architecture
aarch64
Version
Cosmopolitan 3.9.7 MODE=aarch64; #16-Ubuntu SMP PREEMPT_DYNAMIC Sun Sep 21 16:52:46 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2484
tokens/s
33.2
tokens/s
442
ms
pp4096+tg256
1719
tokens/s
30.2
tokens/s
2.42
sec
pp2048+tg256
2153
tokens/s
32.1
tokens/s
981
ms
pp2048+tg768
2182
tokens/s
31.8
tokens/s
973
ms
pp1024+tg1024
2486
tokens/s
33.0
tokens/s
441
ms
pp1280+tg3072
2408
tokens/s
33.3
tokens/s
561
ms
pp384+tg1152
2941
tokens/s
35.6
tokens/s
161
ms
pp64+tg1024
1495
tokens/s
34.4
tokens/s
71
ms
pp16+tg1536
446
tokens/s
35.0
tokens/s
66
ms