TEST #1704 RESULTS
10/26/2025 - 7:43 PM
ACCELERATOR
MODEL
382
tokens/s
generation
57
ms
time to first token
25538
tokens/s
prompt
5562
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen Threadripper PRO 3995WX 64-Cores (znver2)
RAM
251.5GB
OS
Linux
Kernel Release
6.17.5-061705-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #202510231631 SMP PREEMPT_DYNAMIC Thu Oct 23 19:57:43 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
32327
tokens/s
382
tokens/s
34
ms
pp4096+tg256
18843
tokens/s
343
tokens/s
221
ms
pp2048+tg256
27346
tokens/s
364
tokens/s
78
ms
pp2048+tg768
27387
tokens/s
360
tokens/s
78
ms
pp1024+tg1024
35155
tokens/s
383
tokens/s
32
ms
pp1280+tg3072
34095
tokens/s
360
tokens/s
41
ms
pp384+tg1152
37620
tokens/s
409
tokens/s
13
ms
pp64+tg1024
12835
tokens/s
422
tokens/s
7
ms
pp16+tg1536
4235
tokens/s
414
tokens/s
6
ms