TEST #2905 RESULTS
02/26/2026 - 11:37 AM
ACCELERATOR
MODEL
19.7
tokens/s
generation
1.18
sec
time to first token
1156
tokens/s
prompt
268
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
RAM
119.7GB
OS
Linux
Kernel Release
6.17.0-1008-nvidia
Architecture
aarch64
Version
Cosmopolitan 3.9.7 MODE=aarch64; #8-Ubuntu SMP PREEMPT_DYNAMIC Wed Jan 21 17:56:56 UTC 2026
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1425
tokens/s
20.4
tokens/s
769
ms
pp4096+tg256
991
tokens/s
18.1
tokens/s
4.19
sec
pp2048+tg256
1254
tokens/s
19.5
tokens/s
1.69
sec
pp2048+tg768
1252
tokens/s
19.3
tokens/s
1.69
sec
pp1024+tg1024
1428
tokens/s
19.9
tokens/s
767
ms
pp1280+tg3072
1369
tokens/s
18.8
tokens/s
987
ms
pp384+tg1152
1576
tokens/s
20.3
tokens/s
292
ms
pp64+tg1024
874
tokens/s
20.7
tokens/s
158
ms
pp16+tg1536
233
tokens/s
20.5
tokens/s
118
ms