TEST #429 RESULTS
04/06/2025 - 9:29 PM
ACCELERATOR
MODEL
60.3
tokens/s
generation
609
ms
time to first token
2145
tokens/s
prompt
596
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 7 5700X 8-Core Processor (znver3)
RAM
31.9GB
OS
Windows
Kernel Release
10.0
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2530
tokens/s
61.9
tokens/s
422
ms
pp4096+tg256
2007
tokens/s
56.4
tokens/s
2.06
sec
pp2048+tg256
2345
tokens/s
58.7
tokens/s
890
ms
pp2048+tg768
2316
tokens/s
59.4
tokens/s
903
ms
pp1024+tg1024
2557
tokens/s
61.3
tokens/s
417
ms
pp1280+tg3072
2465
tokens/s
58.6
tokens/s
536
ms
pp384+tg1152
2601
tokens/s
62.2
tokens/s
164
ms
pp64+tg1024
1944
tokens/s
63.0
tokens/s
49
ms
pp16+tg1536
537
tokens/s
61.0
tokens/s
45
ms