Inquiry for open source contribution

[githubshaurya]

I want to add MoME into nanoVLM. That requires some (not full) training (of MoME router). Will Hugging Face train the model by themselves? (I don't have enough resources to train it for universal use. Though I can give proof-of-concept and a demo as well) Or my proposal will be rejected? I can edit train.py to enable any individual user to train it on their machine.