TEST #80 RESULTS
03/28/2025 - 9:15 PM
ACCELERATOR
MODEL
58.3
tokens/s
generation
229
ms
time to first token
5925
tokens/s
prompt
1148
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 9 7950X 16-Core Processor (znver4)
RAM
125GB
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
8092
tokens/s
66.4
tokens/s
142
ms
pp4096+tg256
5290
tokens/s
46.0
tokens/s
798
ms
pp2048+tg256
6725
tokens/s
57.8
tokens/s
322
ms
pp2048+tg768
6726
tokens/s
52.3
tokens/s
322
ms
pp1024+tg1024
8157
tokens/s
60.2
tokens/s
141
ms
pp1280+tg3072
7412
tokens/s
51.2
tokens/s
189
ms
pp384+tg1152
7783
tokens/s
63.1
tokens/s
65
ms
pp64+tg1024
2405
tokens/s
64.3
tokens/s
42
ms
pp16+tg1536
736
tokens/s
63.6
tokens/s
37
ms