does vicuna support embedding input?

[lijianxing123]

Does vllm support embedding input in vicuna model, in official version support embedding input like llm_model.generate(inputs_embeds=inputs_embeds)

[WoosukKwon]

@lijianxing123 Thanks for your interest! vLLM currently does not support inputs_embed. The input and output of the model should be in the form of text.

[hujunchao]

Insteresting!

[hmellor]

@WoosukKwon is this a feature you'd like to see implemented?

[hmellor]

Closing in favour of #416