TEST #1718 RESULTS
10/31/2025 - 10:46 PM
ACCELERATOR
MODEL
20.8
tokens/s
generation
3.80
sec
time to first token
418
tokens/s
prompt
132
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 5900X 12-Core Processor (znver3)
RAM
63.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
476
tokens/s
25.9
tokens/s
2.19
sec
pp4096+tg256
273
tokens/s
8.4
tokens/s
15.11
sec
pp2048+tg256
382
tokens/s
18.0
tokens/s
5.41
sec
pp2048+tg768
382
tokens/s
16.5
tokens/s
5.42
sec
pp1024+tg1024
468
tokens/s
21.7
tokens/s
2.23
sec
pp1280+tg3072
448
tokens/s
13.7
tokens/s
2.90
sec
pp384+tg1152
541
tokens/s
26.1
tokens/s
743
ms
pp64+tg1024
497
tokens/s
29.3
tokens/s
158
ms
pp16+tg1536
291
tokens/s
28.0
tokens/s
83
ms