TEST #1645 RESULTS
10/13/2025 - 11:55 AM
ACCELERATOR
MODEL
360
tokens/s
generation
48
ms
time to first token
27505
tokens/s
prompt
5905
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7543 32-Core Processor (znver3)
RAM
944.8GB
OS
Linux
Kernel Release
6.11.0-26-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #26~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Thu Apr 17 19:20:47 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
34590
tokens/s
395
tokens/s
32
ms
pp4096+tg256
24030
tokens/s
304
tokens/s
174
ms
pp2048+tg256
33842
tokens/s
351
tokens/s
63
ms
pp2048+tg768
32968
tokens/s
345
tokens/s
65
ms
pp1024+tg1024
39696
tokens/s
370
tokens/s
28
ms
pp1280+tg3072
34608
tokens/s
337
tokens/s
40
ms
pp384+tg1152
34157
tokens/s
374
tokens/s
14
ms
pp64+tg1024
10414
tokens/s
384
tokens/s
9
ms
pp16+tg1536
3240
tokens/s
376
tokens/s
8
ms