TEST #1303 RESULTS
08/03/2025 - 6:08 PM
ACCELERATOR
MODEL
148
tokens/s
generation
267
ms
time to first token
5276
tokens/s
prompt
1429
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 5800HS with Radeon Graphics (znver3)
RAM
15.4GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
6503
tokens/s
154
tokens/s
164
ms
pp4096+tg256
4240
tokens/s
126
tokens/s
974
ms
pp2048+tg256
5104
tokens/s
146
tokens/s
409
ms
pp2048+tg768
5526
tokens/s
144
tokens/s
378
ms
pp1024+tg1024
6473
tokens/s
151
tokens/s
165
ms
pp1280+tg3072
6126
tokens/s
139
tokens/s
216
ms
pp384+tg1152
7680
tokens/s
154
tokens/s
56
ms
pp64+tg1024
4705
tokens/s
159
tokens/s
20
ms
pp16+tg1536
1130
tokens/s
156
tokens/s
21
ms