TEST #21 RESULTS
03/27/2025 - 7:14 PM
ACCELERATOR
MODEL
82.3
tokens/s
generation
620
ms
time to first token
2475
tokens/s
prompt
690
LocalScore
HOW YOU STACK UP
Explore All ResultsLlama 3.2 1B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7352 24-Core Processor (znver2)
RAM
125.6GB
OS
Linux
Kernel Release
6.8.0-52-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #53~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Wed Jan 15 19:18:46 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
3044
tokens/s
105
tokens/s
346
ms
pp4096+tg256
1679
tokens/s
13.3
tokens/s
2.51
sec
pp2048+tg256
2385
tokens/s
74.5
tokens/s
872
ms
pp2048+tg768
2385
tokens/s
69.6
tokens/s
872
ms
pp1024+tg1024
2998
tokens/s
87.4
tokens/s
351
ms
pp1280+tg3072
2865
tokens/s
47.0
tokens/s
458
ms
pp384+tg1152
3675
tokens/s
105
tokens/s
112
ms
pp64+tg1024
2371
tokens/s
124
tokens/s
34
ms
pp16+tg1536
876
tokens/s
115
tokens/s
25
ms