TEST #3058 RESULTS
03/11/2026 - 9:01 PM
ACCELERATOR
MODEL
44.1
tokens/s
generation
1.96
sec
time to first token
659
tokens/s
prompt
246
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 7B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 5800X3D 8-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
765
tokens/s
51.5
tokens/s
1.36
sec
pp4096+tg256
624
tokens/s
26.8
tokens/s
6.60
sec
pp2048+tg256
713
tokens/s
39.3
tokens/s
2.90
sec
pp2048+tg768
717
tokens/s
37.5
tokens/s
2.88
sec
pp1024+tg1024
763
tokens/s
44.7
tokens/s
1.36
sec
pp1280+tg3072
729
tokens/s
34.8
tokens/s
1.78
sec
pp384+tg1152
765
tokens/s
50.8
tokens/s
518
ms
pp64+tg1024
309
tokens/s
57.6
tokens/s
222
ms
pp16+tg1536
548
tokens/s
53.8
tokens/s
44
ms