TEST #156 RESULTS
03/28/2025 - 11:02 PM
ACCELERATOR
MODEL
43.2
tokens/s
generation
207
ms
time to first token
6307
tokens/s
prompt
1096
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 9554 64-Core Processor (znver4)
RAM
377.4GB
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
8281
tokens/s
47.9
tokens/s
145
ms
pp4096+tg256
6872
tokens/s
37.1
tokens/s
627
ms
pp2048+tg256
7642
tokens/s
44.8
tokens/s
290
ms
pp2048+tg768
7726
tokens/s
40.3
tokens/s
290
ms
pp1024+tg1024
8427
tokens/s
42.6
tokens/s
144
ms
pp1280+tg3072
7691
tokens/s
39.6
tokens/s
190
ms
pp384+tg1152
7699
tokens/s
46.0
tokens/s
68
ms
pp64+tg1024
1870
tokens/s
44.7
tokens/s
59
ms
pp16+tg1536
557
tokens/s
45.5
tokens/s
48
ms