TEST #3509 RESULTS
05/11/2026 - 9:48 PM
ACCELERATOR
MODEL
59.1
tokens/s
generation
674
ms
time to first token
1949
tokens/s
prompt
555
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD EPYC 7251 8-Core Processor
RAM
220.3GB
OS
Linux
Kernel Release
6.17.0-23-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #23~24.04.1-Ubuntu SMP PREEMPT_DYNAMIC Tue Apr 14 16:11:48 UTC 2
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2281
tokens/s
62.3
tokens/s
465
ms
pp4096+tg256
1834
tokens/s
54.3
tokens/s
2.25
sec
pp2048+tg256
2098
tokens/s
58.8
tokens/s
993
ms
pp2048+tg768
2091
tokens/s
57.7
tokens/s
997
ms
pp1024+tg1024
2235
tokens/s
59.7
tokens/s
474
ms
pp1280+tg3072
2144
tokens/s
56.0
tokens/s
614
ms
pp384+tg1152
2377
tokens/s
60.6
tokens/s
178
ms
pp64+tg1024
2000
tokens/s
61.4
tokens/s
48
ms
pp16+tg1536
482
tokens/s
60.9
tokens/s
48
ms