TEST #1137 RESULTS
06/27/2025 - 11:15 AM
MODEL
2.0
tokens/s
generation
243.74
sec
time to first token
6
tokens/s
prompt
4
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
RAM
24.4GB
OS
Linux
Kernel Release
6.8.0-1027-oracle
Architecture
aarch64
Version
Cosmopolitan 3.9.7 MODE=aarch64; #28-Ubuntu SMP Mon May 26 18:47:15 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
5
tokens/s
2.2
tokens/s
214.10
sec
pp4096+tg256
5
tokens/s
1.7
tokens/s
767.99
sec
pp2048+tg256
6
tokens/s
1.9
tokens/s
366.28
sec
pp2048+tg768
6
tokens/s
1.9
tokens/s
364.85
sec
pp1024+tg1024
6
tokens/s
2.0
tokens/s
177.40
sec
pp1280+tg3072
6
tokens/s
1.8
tokens/s
223.93
sec
pp384+tg1152
6
tokens/s
2.0
tokens/s
65.15
sec
pp64+tg1024
6
tokens/s
2.1
tokens/s
10.95
sec
pp16+tg1536
6
tokens/s
2.0
tokens/s
3.03
sec
