Freepik/F-Lite-7B · How much GPU memory does the 7B parameter model use?

Hi team,

Thanks for releasing F Lite 7B, it looks like a very promising lightweight model!

I’d like to get a better idea of its hardware requirements. Could you share details on:

The approximate GPU memory usage when running inference (e.g. batch size 1)

Memory usage during training or fine-tuning, if applicable

Any optimizations (e.g., quantization, reduced precision) that affect memory footprint

This information would really help in planning deployments and evaluating compatibility with various hardware setups.

Thanks in advance!