TEST #1899 RESULTS
12/03/2025 - 1:48 AM
ACCELERATOR
MODEL
3.2
tokens/s
generation
53.53
sec
time to first token
26
tokens/s
prompt
12
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon Gold 5412U (sapphirerapids)
RAM
39.9GB
OS
Linux
Kernel Release
6.8.0-88-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #89-Ubuntu SMP PREEMPT_DYNAMIC Sat Oct 11 01:02:46 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
29
tokens/s
4.5
tokens/s
35.04
sec
pp4096+tg256
24
tokens/s
3.4
tokens/s
167.81
sec
pp2048+tg256
28
tokens/s
3.6
tokens/s
74.17
sec
pp2048+tg768
21
tokens/s
2.4
tokens/s
103.88
sec
pp1024+tg1024
29
tokens/s
3.4
tokens/s
36.11
sec
pp1280+tg3072
27
tokens/s
2.4
tokens/s
47.47
sec
pp384+tg1152
29
tokens/s
3.6
tokens/s
13.54
sec
pp64+tg1024
26
tokens/s
2.5
tokens/s
2.72
sec
pp16+tg1536
20
tokens/s
3.1
tokens/s
1.06
sec