PleIAs/Pleias-Nano · How to make it compatible with the Hugging Face Energy Score leaderboard?

Hi there,

I'm interested in running the Pleias family of models against the Hugging Face Energy Score benchmark tool, that's listed below, but I think I need some pointers on if this is possible, and if so how to do it.

If it helps, I'm refering to this AI Energy Score benchmark and leaderboard:

https://huggingface.co./spaces/AIEnergyScore/2024_Leaderboard

And the source code for the benchmark tool is listed below on github:

https://github.com/huggingface/EnergyStarAI?tab=readme-ov-file

I've tried running the benchmark for both text generation and question Q and A, and the response coming back is:

This model isn't compatible with the chosen task! Pick a different model-task combination

Thanks!

Chris