TEST #1477 RESULTS
09/05/2025 - 12:25 AM
ACCELERATOR
MODEL
57.1
tokens/s
generation
801
ms
time to first token
1593
tokens/s
prompt
484
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 3700X 8-Core Processor (znver2)
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
1874
tokens/s
60.4
tokens/s
563
ms
pp4096+tg256
1556
tokens/s
49.6
tokens/s
2.65
sec
pp2048+tg256
1763
tokens/s
55.8
tokens/s
1.18
sec
pp2048+tg768
1762
tokens/s
55.2
tokens/s
1.18
sec
pp1024+tg1024
1872
tokens/s
58.0
tokens/s
564
ms
pp1280+tg3072
1819
tokens/s
53.1
tokens/s
721
ms
pp384+tg1152
1734
tokens/s
60.1
tokens/s
238
ms
pp64+tg1024
1406
tokens/s
60.8
tokens/s
62
ms
pp16+tg1536
549
tokens/s
60.6
tokens/s
45
ms