TEST #1108 RESULTS
06/25/2025 - 4:17 AM
ACCELERATOR
MODEL
40.3
tokens/s
generation
1.43
sec
time to first token
1028
tokens/s
prompt
307
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 5900X 12-Core Processor (znver3)
RAM
63.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.3
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1238
tokens/s
47.3
tokens/s
850
ms
pp4096+tg256
761
tokens/s
22.7
tokens/s
5.42
sec
pp2048+tg256
1001
tokens/s
35.3
tokens/s
2.08
sec
pp2048+tg768
1007
tokens/s
33.3
tokens/s
2.06
sec
pp1024+tg1024
1200
tokens/s
41.0
tokens/s
876
ms
pp1280+tg3072
1139
tokens/s
29.6
tokens/s
1.15
sec
pp384+tg1152
1392
tokens/s
47.6
tokens/s
295
ms
pp64+tg1024
954
tokens/s
54.6
tokens/s
83
ms
pp16+tg1536
562
tokens/s
51.1
tokens/s
44
ms