TEST #980 RESULTS
06/03/2025 - 3:08 AM
ACCELERATOR
MODEL
77.6
tokens/s
generation
437
ms
time to first token
2918
tokens/s
prompt
804
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct Abliterated - 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
3557
tokens/s
79.6
tokens/s
301
ms
pp4096+tg256
2852
tokens/s
71.6
tokens/s
1.45
sec
pp2048+tg256
3312
tokens/s
78.2
tokens/s
632
ms
pp2048+tg768
3290
tokens/s
74.3
tokens/s
636
ms
pp1024+tg1024
3475
tokens/s
79.6
tokens/s
308
ms
pp1280+tg3072
3335
tokens/s
74.3
tokens/s
396
ms
pp384+tg1152
3348
tokens/s
79.0
tokens/s
127
ms
pp64+tg1024
2627
tokens/s
81.7
tokens/s
36
ms
pp16+tg1536
468
tokens/s
80.3
tokens/s
45
ms