TEST #3682 RESULTS
06/03/2026 - 3:48 PM
ACCELERATOR
MODEL
20.0
tokens/s
generation
1.16
sec
time to first token
1174
tokens/s
prompt
272
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
RAM
121.6GB
OS
Linux
Kernel Release
6.17.0-1018-nvidia
Architecture
aarch64
Version
Cosmopolitan 3.9.7 MODE=aarch64; #18-Ubuntu SMP PREEMPT_DYNAMIC Tue May 5 21:28:33 UTC 2026
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1459
tokens/s
20.6
tokens/s
752
ms
pp4096+tg256
1014
tokens/s
18.2
tokens/s
4.10
sec
pp2048+tg256
1274
tokens/s
19.7
tokens/s
1.66
sec
pp2048+tg768
1270
tokens/s
19.5
tokens/s
1.66
sec
pp1024+tg1024
1449
tokens/s
20.3
tokens/s
758
ms
pp1280+tg3072
1398
tokens/s
19.1
tokens/s
966
ms
pp384+tg1152
1604
tokens/s
20.8
tokens/s
288
ms
pp64+tg1024
894
tokens/s
21.1
tokens/s
164
ms
pp16+tg1536
203
tokens/s
20.9
tokens/s
126
ms