Text Generation
Transformers
PyTorch
mistral
openchat
C-RLFT
conversational
Inference Endpoints
text-generation-inference

Request for Model Sharding of OpenChat 3.5 for Improved Accessibility

#10
by Firejowl - opened

Hello OpenChat Team,

I hope this message finds you well. I am reaching out to express my appreciation for the OpenChat 3.5 model and to discuss a suggestion that could potentially extend its reach and usability.

One of the challenges I have encountered is the limitation posed by hardware requirements, which can be a barrier for individuals or researchers with access to only lower-end computers. To address this, I am interested in exploring the possibility of model sharding for OpenChat 3.5.

Sharding the model could democratize access, allowing users with less powerful hardware to experiment with and benefit from the model's capabilities. This approach could also optimize the use of the model on popular cloud platforms like Google Colab and Kaggle, which are essential tools for many in the AI community.

The technical intricacies of model sharding notwithstanding, its implementation could lead to broader adoption and spur innovation by enabling a wider audience to work with OpenChat 3.5.

I would love to hear your thoughts on this proposal or any alternative solutions that could help mitigate the hardware constraints faced by many users.

Thank you for your contributions to the field and for considering this enhancement.

+1
A free colab notebook tutorial on how to run the model will be amazing :)

Sign up or log in to comment