TEST #991 RESULTS
06/04/2025 - 9:46 AM
ACCELERATOR
MODEL
52.2
tokens/s
generation
738
ms
time to first token
1719
tokens/s
prompt
496
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-2697A v4 @ 2.60GHz (broadwell)
RAM
125.7GB
OS
Linux
Kernel Release
6.1.0-34-amd64
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC Debian 6.1.135-1 (2025-04-25)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2047
tokens/s
54.9
tokens/s
518
ms
pp4096+tg256
1710
tokens/s
47.3
tokens/s
2.42
sec
pp2048+tg256
1918
tokens/s
51.6
tokens/s
1.09
sec
pp2048+tg768
1911
tokens/s
50.5
tokens/s
1.09
sec
pp1024+tg1024
2024
tokens/s
52.8
tokens/s
525
ms
pp1280+tg3072
1999
tokens/s
49.6
tokens/s
659
ms
pp384+tg1152
2047
tokens/s
54.1
tokens/s
227
ms
pp64+tg1024
1405
tokens/s
55.0
tokens/s
63
ms
pp16+tg1536
413
tokens/s
54.2
tokens/s
58
ms