TEST #1330 RESULTS
08/07/2025 - 7:57 AM
ACCELERATOR
MODEL
24.6
tokens/s
generation
2.55
sec
time to first token
542
tokens/s
prompt
174
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen3 4B Thinking 2507 - Q4_K - Medium
SYSTEM
CPU
Intel Core i5-6600K CPU @ 3.50GHz (skylake)
RAM
15.1GB
OS
Linux
Kernel Release
6.14.0-27-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #27-Ubuntu SMP PREEMPT_DYNAMIC Tue Jul 22 17:01:58 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
609
tokens/s
28.4
tokens/s
1.72
sec
pp4096+tg256
466
tokens/s
15.4
tokens/s
8.86
sec
pp2048+tg256
549
tokens/s
22.1
tokens/s
3.77
sec
pp2048+tg768
549
tokens/s
20.9
tokens/s
3.78
sec
pp1024+tg1024
602
tokens/s
25.1
tokens/s
1.74
sec
pp1280+tg3072
582
tokens/s
19.3
tokens/s
2.24
sec
pp384+tg1152
643
tokens/s
28.7
tokens/s
628
ms
pp64+tg1024
584
tokens/s
31.7
tokens/s
137
ms
pp16+tg1536
293
tokens/s
29.9
tokens/s
81
ms