Is it a good idea to utilize the multimodal input and output capabilities of BaseChatModel to support text-to-image/audio/video models, such as speech-to-text models, text-to-speech models, Stable Diffusion or Flux models?

[zhcn000000]

Checked

• I searched existing ideas and did not find a similar one
• I added a very descriptive title
• I've clearly described the feature request and motivation for it

Feature request

Is it a good idea to utilize the multimodal input and output capabilities of BaseChatModel to support text-to-image/audio/video models, such as speech-to-text models, text-to-speech models, Stable Diffusion or Flux models?
For example, my own repository: langchain-multimedia

Motivation

This enables a unified interface to support text-to-image, text-to-audio, and text-to-video models.

Proposal (If applicable)

No response