[Realtime API] Adds minimal realtime API based on websockets (#33187)

Signed-off-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: Nick Hill <nickhill123@gmail.com>
This commit is contained in:
Patrick von Platen
2026-01-30 11:41:29 +01:00
committed by GitHub
parent 1a7894dbdf
commit 10152d2194
21 changed files with 1316 additions and 48 deletions

View File

@@ -2,7 +2,7 @@
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
from typing import Literal, get_args
GenerationTask = Literal["generate", "transcription"]
GenerationTask = Literal["generate", "transcription", "realtime"]
GENERATION_TASKS: tuple[GenerationTask, ...] = get_args(GenerationTask)
PoolingTask = Literal[