TEST #2017 RESULTS
12/21/2025 - 5:09 PM
ACCELERATOR
MODEL
30.5
tokens/s
generation
3.05
sec
time to first token
1245
tokens/s
prompt
232
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 5600GT with Radeon Graphics (znver3)
RAM
31.1GB
OS
Linux
Kernel Release
6.17.9-76061709-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #202511241048~1764607909~24.04~df6b2b6 SMP PREEMPT_DYNAMIC Mon D
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1529
tokens/s
33.4
tokens/s
700
ms
pp4096+tg256
1098
tokens/s
23.7
tokens/s
3.77
sec
pp2048+tg256
1345
tokens/s
29.1
tokens/s
1.56
sec
pp2048+tg768
1344
tokens/s
28.0
tokens/s
1.56
sec
pp1024+tg1024
1481
tokens/s
30.5
tokens/s
722
ms
pp1280+tg3072
1451
tokens/s
26.7
tokens/s
914
ms
pp384+tg1152
1680
tokens/s
33.2
tokens/s
257
ms
pp64+tg1024
1272
tokens/s
35.2
tokens/s
78
ms
pp16+tg1536
1
tokens/s
34.5
tokens/s
17.90
sec