Skip to content

vllm_omni.entrypoints.duplex.openai ¶

logger `module-attribute` ¶

logger = init_logger(__name__)

dispatch_generic_realtime_websocket `async` ¶

dispatch_generic_realtime_websocket(
    websocket: WebSocket,
) -> None

Fall back to the generic OpenAI Realtime API for non-Qwen3-Omni models.

dispatch_realtime_websocket `async` ¶

dispatch_realtime_websocket(websocket: WebSocket) -> None

Handle Realtime sessions not already routed to the proprietary duplex handler by the caller: Qwen3-Omni gets a conformant full-duplex connection, everything else falls back to the generic turn-based Realtime API.

reject_realtime_websocket `async` ¶

reject_realtime_websocket(
    websocket: WebSocket, message: str
) -> None

supports_qwen3_omni_realtime ¶

supports_qwen3_omni_realtime(
    stage_configs: Sequence[BaseVllmOmniStageConfig] | None,
) -> bool