TEST #800 RESULTS
05/03/2025 - 3:01 PM
ACCELERATOR
MODEL
49.7
tokens/s
generation
933
ms
time to first token
1400
tokens/s
prompt
421
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 5950X 16-Core Processor (znver3)
RAM
125.7GB
OS
Linux
Kernel Release
6.11.0-112021-tuxedo
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #21~24.04.1tux1 SMP PREEMPT_DYNAMIC Tue Mar 11 20:03:05 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1662
tokens/s
53.4
tokens/s
635
ms
pp4096+tg256
1331
tokens/s
44.4
tokens/s
3.10
sec
pp2048+tg256
1479
tokens/s
48.9
tokens/s
1.40
sec
pp2048+tg768
1486
tokens/s
47.9
tokens/s
1.40
sec
pp1024+tg1024
1615
tokens/s
50.5
tokens/s
653
ms
pp1280+tg3072
1556
tokens/s
46.7
tokens/s
842
ms
pp384+tg1152
1673
tokens/s
51.3
tokens/s
249
ms
pp64+tg1024
1341
tokens/s
52.4
tokens/s
66
ms
pp16+tg1536
458
tokens/s
51.8
tokens/s
53
ms