TEST #79 RESULTS
03/28/2025 - 9:14 PM
ACCELERATOR
MODEL
56.5
tokens/s
generation
518
ms
time to first token
2617
tokens/s
prompt
658
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7352 24-Core Processor (znver2)
RAM
251.6GB
OS
Linux
Kernel Release
6.8.0-40-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #40~22.04.3-Ubuntu SMP PREEMPT_DYNAMIC Tue Jul 30 17:30:19 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
3144
tokens/s
57.9
tokens/s
343
ms
pp4096+tg256
2288
tokens/s
52.7
tokens/s
1.81
sec
pp2048+tg256
2788
tokens/s
56.0
tokens/s
753
ms
pp2048+tg768
2789
tokens/s
55.6
tokens/s
752
ms
pp1024+tg1024
3121
tokens/s
57.1
tokens/s
346
ms
pp1280+tg3072
3018
tokens/s
54.6
tokens/s
442
ms
pp384+tg1152
3543
tokens/s
57.8
tokens/s
126
ms
pp64+tg1024
2351
tokens/s
58.6
tokens/s
44
ms
pp16+tg1536
510
tokens/s
58.3
tokens/s
48
ms