TEST #1811 RESULTS
11/16/2025 - 10:54 AM
ACCELERATOR
MODEL
59.7
tokens/s
generation
696
ms
time to first token
1745
tokens/s
prompt
531
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 5600X 6-Core Processor (znver3)
RAM
31.9GB
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
1433
tokens/s
61.2
tokens/s
731
ms
pp4096+tg256
1864
tokens/s
56.0
tokens/s
2.22
sec
pp2048+tg256
2182
tokens/s
59.5
tokens/s
955
ms
pp2048+tg768
2158
tokens/s
58.6
tokens/s
967
ms
pp1024+tg1024
2242
tokens/s
60.8
tokens/s
473
ms
pp1280+tg3072
2249
tokens/s
57.7
tokens/s
586
ms
pp384+tg1152
1952
tokens/s
61.0
tokens/s
214
ms
pp64+tg1024
1029
tokens/s
61.7
tokens/s
78
ms
pp16+tg1536
594
tokens/s
61.1
tokens/s
43
ms