TEST #1517 RESULTS
09/14/2025 - 10:35 AM
ACCELERATOR
MODEL
45.7
tokens/s
generation
123
ms
time to first token
11191
tokens/s
prompt
1610
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 9950X3D 16-Core Processor
RAM
63.6GB
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
14750
tokens/s
45.6
tokens/s
92
ms
pp4096+tg256
13290
tokens/s
44.3
tokens/s
332
ms
pp2048+tg256
14484
tokens/s
45.6
tokens/s
163
ms
pp2048+tg768
15383
tokens/s
45.5
tokens/s
155
ms
pp1024+tg1024
15176
tokens/s
45.7
tokens/s
89
ms
pp1280+tg3072
13347
tokens/s
45.2
tokens/s
118
ms
pp384+tg1152
11778
tokens/s
46.1
tokens/s
55
ms
pp64+tg1024
1983
tokens/s
46.8
tokens/s
51
ms
pp16+tg1536
527
tokens/s
46.5
tokens/s
49
ms