TEST #3614 RESULTS
05/26/2026 - 11:06 AM
ACCELERATOR
MODEL
26.6
tokens/s
generation
2.25
sec
time to first token
613
tokens/s
prompt
193
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 5700X 8-Core Processor (znver3)
RAM
30.3GB
OS
Linux
Kernel Release
7.0.0-15-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #15-Ubuntu SMP PREEMPT_DYNAMIC Wed Apr 22 16:06:43 UTC 2026
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
684
tokens/s
30.7
tokens/s
1.53
sec
pp4096+tg256
530
tokens/s
16.2
tokens/s
7.79
sec
pp2048+tg256
624
tokens/s
23.7
tokens/s
3.32
sec
pp2048+tg768
621
tokens/s
22.6
tokens/s
3.34
sec
pp1024+tg1024
680
tokens/s
27.0
tokens/s
1.54
sec
pp1280+tg3072
656
tokens/s
20.8
tokens/s
1.99
sec
pp384+tg1152
704
tokens/s
31.1
tokens/s
574
ms
pp64+tg1024
685
tokens/s
34.6
tokens/s
119
ms
pp16+tg1536
333
tokens/s
32.3
tokens/s
73
ms