TEST #764 RESULTS
04/24/2025 - 11:06 PM
ACCELERATOR
MODEL
76.6
tokens/s
generation
421
ms
time to first token
3028
tokens/s
prompt
820
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 5800X 8-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
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
3673
tokens/s
77.4
tokens/s
291
ms
pp4096+tg256
2964
tokens/s
72.9
tokens/s
1.40
sec
pp2048+tg256
3394
tokens/s
76.4
tokens/s
617
ms
pp2048+tg768
3387
tokens/s
76.0
tokens/s
618
ms
pp1024+tg1024
3658
tokens/s
77.5
tokens/s
293
ms
pp1280+tg3072
3487
tokens/s
75.0
tokens/s
380
ms
pp384+tg1152
3689
tokens/s
78.2
tokens/s
119
ms
pp64+tg1024
2334
tokens/s
78.0
tokens/s
40
ms
pp16+tg1536
668
tokens/s
78.2
tokens/s
37
ms