TEST #347 RESULTS
04/06/2025 - 2:39 AM
ACCELERATOR
MODEL
25.7
tokens/s
generation
448
ms
time to first token
2862
tokens/s
prompt
548
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 7800X3D 8-Core Processor (znver4)
RAM
63.1GB
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
3782
tokens/s
27.9
tokens/s
306
ms
pp4096+tg256
3170
tokens/s
23.5
tokens/s
1.34
sec
pp2048+tg256
3445
tokens/s
25.7
tokens/s
632
ms
pp2048+tg768
3503
tokens/s
25.5
tokens/s
622
ms
pp1024+tg1024
3821
tokens/s
26.6
tokens/s
306
ms
pp1280+tg3072
3544
tokens/s
25.0
tokens/s
398
ms
pp384+tg1152
3678
tokens/s
26.2
tokens/s
147
ms
pp64+tg1024
639
tokens/s
25.4
tokens/s
147
ms
pp16+tg1536
173
tokens/s
25.5
tokens/s
134
ms