TEST #1684 RESULTS
10/18/2025 - 9:52 PM
ACCELERATOR
MODEL
5.2
tokens/s
generation
1.47
sec
time to first token
1122
tokens/s
prompt
158
LocalScore
HOW YOU STACK UP
Explore All ResultsDeepSeek R1 0528 Qwen3 8B - Q8_0
SYSTEM
CPU
AMD Ryzen 7 7700 8-Core Processor (znver4)
RAM
63.1GB
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
1485
tokens/s
5.6
tokens/s
868
ms
pp4096+tg256
815
tokens/s
1.7
tokens/s
5.60
sec
pp2048+tg256
1175
tokens/s
3.8
tokens/s
2.04
sec
pp2048+tg768
1182
tokens/s
3.5
tokens/s
2.03
sec
pp1024+tg1024
1484
tokens/s
5.0
tokens/s
872
ms
pp1280+tg3072
1332
tokens/s
2.9
tokens/s
1.17
sec
pp384+tg1152
1609
tokens/s
6.9
tokens/s
343
ms
pp64+tg1024
800
tokens/s
9.4
tokens/s
144
ms
pp16+tg1536
212
tokens/s
7.9
tokens/s
130
ms