[Feature] Add Lingen - MATE architecture (SSM Mamba2 based) as a model option

[mgyong]

Describe the feature

LinGen: Towards High-Resolution Minute-Length Text-to-Video Generation
with Linear Computational Complexity
https://arxiv.org/pdf/2412.09856

LinGen is non-transformer based SSM Mamba2 based model. We are working with the first author to implement this LinGen MATE architecture for video generation. We like to add LinGen MATE architecture as a model option to InternLM EVO. We would like to discuss with internlm EVO team about this implementation

Will you implement it?

• I would like to implement this feature and create a PR!

[zigzagcai]

Hi, @mgyong

Very glad to see that you like to integrate LinGen as model option into our framework.
Also, I have some experience with SSM Mamba, so I will follow up on this issue!

[mgyong]

@zigzagcai great can we communicate on wechat as well 18510125703 thanks
We have some questions on how to integrate Lingen into internlm. Thanks