TEST #1091 RESULTS
06/23/2025 - 9:14 AM
ACCELERATOR
MODEL
83.1
tokens/s
generation
347
ms
time to first token
3641
tokens/s
prompt
955
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 5900X 12-Core Processor (znver3)
RAM
53.6GB
OS
Linux
Kernel Release
5.15.167.4-microsoft-standard-WSL2
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP Tue Nov 5 00:21:55 UTC 2024
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
4350
tokens/s
85.6
tokens/s
247
ms
pp4096+tg256
3674
tokens/s
83.4
tokens/s
1.13
sec
pp2048+tg256
4142
tokens/s
82.6
tokens/s
507
ms
pp2048+tg768
4150
tokens/s
83.5
tokens/s
506
ms
pp1024+tg1024
4380
tokens/s
78.7
tokens/s
246
ms
pp1280+tg3072
4149
tokens/s
81.8
tokens/s
321
ms
pp384+tg1152
4470
tokens/s
83.9
tokens/s
98
ms
pp64+tg1024
2788
tokens/s
84.5
tokens/s
35
ms
pp16+tg1536
669
tokens/s
83.9
tokens/s
35
ms