[New Model]: glm-4-voice-9b

[Luffy-ZY-Wang]

The model to consider.

huggingface link: https://huggingface.co/THUDM/glm-4-voice-9b
github link: https://github.com/THUDM/GLM-4-Voice?tab=readme-ov-file

The closest model vllm already supports.

whisper, glm-4

btw, this model is actually whisper encoder + glm-4-9b + CosyVoice

What's your difficulty of supporting the model you want?

This is an end to end large audio model. It may possess a new model architecture. Please see its repo or report paper for details.

Before submitting a new issue...

• Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.

[LilLeapo]

Can this model use vllm?

[Luffy-ZY-Wang]

Currently vllm doesn't support this model. It's still yet to be done. See Multi-modal Model Requests