TEST #753 RESULTS
04/22/2025 - 12:00 AM
ACCELERATOR
MODEL
103
tokens/s
generation
96
ms
time to first token
13606
tokens/s
prompt
2447
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 9950X 16-Core Processor
RAM
60.4GB
OS
Linux
Kernel Release
6.14.0-061400-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #202503241442 SMP PREEMPT_DYNAMIC Mon Mar 24 15:06:01 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
18213
tokens/s
124
tokens/s
64
ms
pp4096+tg256
14226
tokens/s
84.0
tokens/s
301
ms
pp2048+tg256
16478
tokens/s
101
tokens/s
134
ms
pp2048+tg768
16463
tokens/s
95.7
tokens/s
134
ms
pp1024+tg1024
18955
tokens/s
102
tokens/s
63
ms
pp1280+tg3072
15980
tokens/s
92.1
tokens/s
90
ms
pp384+tg1152
17290
tokens/s
109
tokens/s
31
ms
pp64+tg1024
3812
tokens/s
113
tokens/s
25
ms
pp16+tg1536
1037
tokens/s
111
tokens/s
23
ms