TEST #125 RESULTS
03/28/2025 - 10:07 PM
ACCELERATOR
MODEL
83.2
tokens/s
generation
520
ms
time to first token
2509
tokens/s
prompt
738
LocalScore
HOW YOU STACK UP
Explore All ResultsMeta Llama 3.1 8B Instruct - Q4_K - Medium
SYSTEM
CPU
AMD Ryzen 5 5500 (znver3)
RAM
86.3GB
OS
Linux
Kernel Release
5.15.0-126-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #136~20.04.1-Ubuntu SMP Thu Nov 14 16:38:05 UTC 2024
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
2939
tokens/s
86.6
tokens/s
360
ms
pp4096+tg256
2363
tokens/s
75.1
tokens/s
1.75
sec
pp2048+tg256
2709
tokens/s
81.8
tokens/s
768
ms
pp2048+tg768
2707
tokens/s
80.9
tokens/s
769
ms
pp1024+tg1024
2925
tokens/s
84.0
tokens/s
362
ms
pp1280+tg3072
2838
tokens/s
79.2
tokens/s
463
ms
pp384+tg1152
3174
tokens/s
86.4
tokens/s
132
ms
pp64+tg1024
2312
tokens/s
87.6
tokens/s
39
ms
pp16+tg1536
617
tokens/s
86.8
tokens/s
37
ms