TEST #3895 RESULTS
07/12/2026 - 1:30 PM
ACCELERATOR
MODEL
4.1
tokens/s
generation
37.18
sec
time to first token
38
tokens/s
prompt
16
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen AI 9 HX 370 w/ Radeon 890M
RAM
30GB
OS
Linux
Kernel Release
7.0.0-22-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #22-Ubuntu SMP PREEMPT_DYNAMIC Mon May 25 15:54:34 UTC 2026
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
39
tokens/s
4.2
tokens/s
26.61
sec
pp4096+tg256
34
tokens/s
3.8
tokens/s
119.66
sec
pp2048+tg256
37
tokens/s
4.0
tokens/s
55.75
sec
pp2048+tg768
37
tokens/s
4.0
tokens/s
55.38
sec
pp1024+tg1024
39
tokens/s
4.1
tokens/s
26.48
sec
pp1280+tg3072
33
tokens/s
3.9
tokens/s
38.64
sec
pp384+tg1152
40
tokens/s
4.2
tokens/s
9.74
sec
pp64+tg1024
43
tokens/s
4.3
tokens/s
1.71
sec
pp16+tg1536
37
tokens/s
4.2
tokens/s
661
ms