TEST #3389 RESULTS
04/26/2026 - 3:29 PM
ACCELERATOR
MODEL
115
tokens/s
generation
538
ms
time to first token
2244
tokens/s
prompt
783
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon CPU E5-2680 v4 @ 2.40GHz (broadwell)
RAM
31.2GB
OS
Linux
Kernel Release
6.17.13-3-pve
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC PMX 6.17.13-3 (2026-03-31T21:50Z)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2793
tokens/s
126
tokens/s
375
ms
pp4096+tg256
2390
tokens/s
81.2
tokens/s
1.73
sec
pp2048+tg256
2636
tokens/s
106
tokens/s
787
ms
pp2048+tg768
2491
tokens/s
103
tokens/s
832
ms
pp1024+tg1024
2767
tokens/s
118
tokens/s
378
ms
pp1280+tg3072
2642
tokens/s
97.2
tokens/s
493
ms
pp384+tg1152
2739
tokens/s
130
tokens/s
147
ms
pp64+tg1024
1326
tokens/s
140
tokens/s
55
ms
pp16+tg1536
410
tokens/s
134
tokens/s
45
ms