TEST #684 RESULTS
04/10/2025 - 7:39 PM
ACCELERATOR
MODEL
416
tokens/s
generation
53
ms
time to first token
27892
tokens/s
prompt
6030
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen Threadripper PRO 5975WX 32-Cores (znver3)
RAM
503.5GB
OS
Linux
Kernel Release
6.8.0-52-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #53-Ubuntu SMP PREEMPT_DYNAMIC Sat Jan 11 00:06:25 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
38338
tokens/s
414
tokens/s
29
ms
pp4096+tg256
19619
tokens/s
370
tokens/s
212
ms
pp2048+tg256
29198
tokens/s
396
tokens/s
73
ms
pp2048+tg768
29135
tokens/s
400
tokens/s
73
ms
pp1024+tg1024
40132
tokens/s
420
tokens/s
28
ms
pp1280+tg3072
37007
tokens/s
395
tokens/s
37
ms
pp384+tg1152
39267
tokens/s
442
tokens/s
12
ms
pp64+tg1024
13872
tokens/s
459
tokens/s
7
ms
pp16+tg1536
4460
tokens/s
448
tokens/s
6
ms