TEST #123 RESULTS
03/28/2025 - 10:06 PM
ACCELERATOR
MODEL
53.1
tokens/s
generation
885
ms
time to first token
1487
tokens/s
prompt
447
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 5600X 6-Core Processor (znver3)
RAM
31.3GB
OS
Linux
Kernel Release
5.15.0-131-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #141~20.04.1-Ubuntu SMP Thu Jan 16 18:38:51 UTC 2025
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1531
tokens/s
55.0
tokens/s
687
ms
pp4096+tg256
1369
tokens/s
46.8
tokens/s
3.01
sec
pp2048+tg256
1622
tokens/s
52.1
tokens/s
1.28
sec
pp2048+tg768
1623
tokens/s
51.4
tokens/s
1.28
sec
pp1024+tg1024
1768
tokens/s
54.0
tokens/s
597
ms
pp1280+tg3072
1714
tokens/s
50.2
tokens/s
765
ms
pp384+tg1152
1857
tokens/s
55.5
tokens/s
225
ms
pp64+tg1024
1530
tokens/s
56.8
tokens/s
59
ms
pp16+tg1536
370
tokens/s
56.0
tokens/s
60
ms