sgl-project/sglang

[Feature] Support `return_hidden_states` in async engine

Fechada

#6.528 aberto em 22 de mai. de 2025

 (1 comentário) (0 reação) (0 responsável)Python (6.216 forks)auto 404
good first issue

Métricas do repositório

Stars
 (28.442 estrelas)
Métricas de merge de PR
 (Mesclagem média 1d 13h) (1.000 fundiu PRs em 30d)

Description

Checklist

Motivation

Currently, it is possible to get hidden states using the offline interface, i.e. Engine.generate, but Engine.async_generate doesn't support this function.

result = await model.async_generate(
        ["hello", "how are you doing?"],
        sampling_params={"temperature": 0, "max_new_tokens": 1},
        return_hidden_states=True)
TypeError: Engine.async_generate() got an unexpected keyword argument 'return_hidden_states'

This can be useful when developing web services where some additional hidden states postprocess logics is done.

Related resources

No response

Guia do colaborador