TEST #639 RESULTS
04/09/2025 - 4:00 PM
ACCELERATOR
MODEL
12.1
tokens/s
generation
557
ms
time to first token
3324
tokens/s
prompt
417
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 7700X 8-Core Processor (znver4)
RAM
31.6GB
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
4647
tokens/s
12.9
tokens/s
297
ms
pp4096+tg256
2017
tokens/s
3.7
tokens/s
2.29
sec
pp2048+tg256
3301
tokens/s
8.5
tokens/s
756
ms
pp2048+tg768
3333
tokens/s
7.6
tokens/s
750
ms
pp1024+tg1024
4636
tokens/s
11.3
tokens/s
299
ms
pp1280+tg3072
3945
tokens/s
6.4
tokens/s
416
ms
pp384+tg1152
5082
tokens/s
16.3
tokens/s
116
ms
pp64+tg1024
2238
tokens/s
23.4
tokens/s
49
ms
pp16+tg1536
720
tokens/s
19.0
tokens/s
39
ms