TEST #3441 RESULTS
05/04/2026 - 6:18 PM
ACCELERATOR
MODEL
76.4
tokens/s
generation
410
ms
time to first token
3128
tokens/s
prompt
835
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
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
3823
tokens/s
78.2
tokens/s
281
ms
pp4096+tg256
3006
tokens/s
73.4
tokens/s
1.38
sec
pp2048+tg256
3514
tokens/s
76.0
tokens/s
596
ms
pp2048+tg768
3494
tokens/s
75.6
tokens/s
599
ms
pp1024+tg1024
3799
tokens/s
77.3
tokens/s
282
ms
pp1280+tg3072
3629
tokens/s
75.1
tokens/s
366
ms
pp384+tg1152
3872
tokens/s
77.4
tokens/s
113
ms
pp64+tg1024
2324
tokens/s
77.5
tokens/s
40
ms
pp16+tg1536
687
tokens/s
77.2
tokens/s
37
ms