TEST #601 RESULTS
04/08/2025 - 12:43 PM
ACCELERATOR
MODEL
82.6
tokens/s
generation
1.21
sec
time to first token
1046
tokens/s
prompt
415
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 3B Instruct - Q3_K - Large
SYSTEM
CPU
AMD Ryzen 9 5950X 16-Core Processor (znver3)
RAM
62.7GB
OS
Linux
Kernel Release
6.14.1-1-cachyos
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Tue, 08 Apr 2025 06:33:52 +0000
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1188
tokens/s
97.3
tokens/s
872
ms
pp4096+tg256
1054
tokens/s
45.5
tokens/s
3.91
sec
pp2048+tg256
1135
tokens/s
69.1
tokens/s
1.82
sec
pp2048+tg768
1139
tokens/s
64.8
tokens/s
1.81
sec
pp1024+tg1024
1185
tokens/s
82.5
tokens/s
875
ms
pp1280+tg3072
1168
tokens/s
59.5
tokens/s
1.11
sec
pp384+tg1152
1175
tokens/s
100
tokens/s
335
ms
pp64+tg1024
721
tokens/s
117
tokens/s
95
ms
pp16+tg1536
650
tokens/s
107
tokens/s
31
ms