TEST #868 RESULTS
05/14/2025 - 4:46 PM
ACCELERATOR
MODEL
117
tokens/s
generation
298
ms
time to first token
4467
tokens/s
prompt
1205
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen3-4B - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7352 24-Core Processor (znver2)
RAM
125.6GB
OS
Linux
Kernel Release
6.8.0-59-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #61~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Tue Apr 15 17:03:15 UTC 2
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
5515
tokens/s
116
tokens/s
194
ms
pp4096+tg256
3862
tokens/s
102
tokens/s
1.07
sec
pp2048+tg256
4998
tokens/s
117
tokens/s
419
ms
pp2048+tg768
5040
tokens/s
113
tokens/s
416
ms
pp1024+tg1024
5425
tokens/s
116
tokens/s
198
ms
pp1280+tg3072
5309
tokens/s
112
tokens/s
249
ms
pp384+tg1152
5796
tokens/s
123
tokens/s
74
ms
pp64+tg1024
3775
tokens/s
128
tokens/s
24
ms
pp16+tg1536
480
tokens/s
125
tokens/s
41
ms