TEST #3159 RESULTS
03/23/2026 - 10:34 AM
ACCELERATOR
MODEL
63.4
tokens/s
generation
877
ms
time to first token
1452
tokens/s
prompt
472
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon W-2135 CPU @ 3.70GHz (skylake-avx512)
RAM
125.4GB
OS
Linux
Kernel Release
6.17.13-2-pve
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #1 SMP PREEMPT_DYNAMIC PMX 6.17.13-2 (2026-03-13T08:06Z)
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1772
tokens/s
59.6
tokens/s
595
ms
pp4096+tg256
1377
tokens/s
42.2
tokens/s
3.00
sec
pp2048+tg256
1610
tokens/s
57.8
tokens/s
1.29
sec
pp2048+tg768
1612
tokens/s
56.4
tokens/s
1.29
sec
pp1024+tg1024
1770
tokens/s
66.6
tokens/s
592
ms
pp1280+tg3072
1648
tokens/s
53.1
tokens/s
791
ms
pp384+tg1152
1836
tokens/s
74.8
tokens/s
221
ms
pp64+tg1024
1028
tokens/s
81.8
tokens/s
73
ms
pp16+tg1536
412
tokens/s
78.4
tokens/s
49
ms