Skip to content

Add support for the Mamba2 model. #481

@donpellegrino

Description

@donpellegrino

The Mamba2 model is used modularly in other models, such as IBM's Granite4, and would be useful to have available in rust-bert. For HuggingFace SafeTensors compatible weights see https://huggingface.co/AntonV/mamba2-130m-hf. See https://arxiv.org/pdf/[2405.21060](https://arxiv.org/pdf/2405.21060) for detilas on Mamba2. See https://www.ibm.com/new/announcements/ibm-granite-4-0-tiny-preview-sneak-peek for information on its use in another model.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions