TEST #20 RESULTS
03/26/2025 - 9:07 PM
ACCELERATOR
MODEL
26.8
tokens/s
generation
2.58
sec
time to first token
566
tokens/s
prompt
180
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5.1 Coder 7B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7352 24-Core Processor (znver2)
RAM
125.64GB
OS
Linux
Kernel Release
6.8.0-52-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #53~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Wed Jan 15 19:18:46 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
637
tokens/s
32.9
tokens/s
1.64
sec
pp4096+tg256
431
tokens/s
5.5
tokens/s
9.68
sec
pp2048+tg256
559
tokens/s
26.1
tokens/s
3.70
sec
pp2048+tg768
559
tokens/s
24.9
tokens/s
3.70
sec
pp1024+tg1024
636
tokens/s
29.5
tokens/s
1.64
sec
pp1280+tg3072
617
tokens/s
17.9
tokens/s
2.11
sec
pp384+tg1152
689
tokens/s
33.2
tokens/s
583
ms
pp64+tg1024
595
tokens/s
36.3
tokens/s
132
ms
pp16+tg1536
369
tokens/s
34.8
tokens/s
68
ms