TEST #1534 RESULTS
09/20/2025 - 12:26 AM
ACCELERATOR
MODEL
65.4
tokens/s
generation
2.36
sec
time to first token
2251
tokens/s
prompt
396
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
127.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2585
tokens/s
68.9
tokens/s
412
ms
pp4096+tg256
2245
tokens/s
61.9
tokens/s
1.84
sec
pp2048+tg256
2664
tokens/s
65.8
tokens/s
785
ms
pp2048+tg768
2531
tokens/s
66.1
tokens/s
834
ms
pp1024+tg1024
2820
tokens/s
67.5
tokens/s
379
ms
pp1280+tg3072
2715
tokens/s
61.8
tokens/s
486
ms
pp384+tg1152
2438
tokens/s
64.4
tokens/s
174
ms
pp64+tg1024
2262
tokens/s
65.5
tokens/s
44
ms
pp16+tg1536
1
tokens/s
66.4
tokens/s
16.30
sec