TEST #1777 RESULTS
11/10/2025 - 7:55 PM
ACCELERATOR
MODEL
27.0
tokens/s
generation
1.94
sec
time to first token
579
tokens/s
prompt
200
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen3-30B-A3B-Instruct-2507 - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 5600X 6-Core Processor (znver3)
RAM
63.9GB
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
743
tokens/s
27.8
tokens/s
1.41
sec
pp4096+tg256
713
tokens/s
26.5
tokens/s
5.78
sec
pp2048+tg256
741
tokens/s
27.3
tokens/s
2.80
sec
pp2048+tg768
745
tokens/s
26.4
tokens/s
2.79
sec
pp1024+tg1024
735
tokens/s
27.3
tokens/s
1.43
sec
pp1280+tg3072
680
tokens/s
26.2
tokens/s
1.92
sec
pp384+tg1152
612
tokens/s
27.5
tokens/s
664
ms
pp64+tg1024
180
tokens/s
27.1
tokens/s
393
ms
pp16+tg1536
60
tokens/s
26.5
tokens/s
309
ms