Skip to content

vllm_omni.model_executor.models.personaplex.duplex.plugin

PersonaPlex full-duplex model plugin: engine policy and session policy in one class.

PersonaPlex is a pure-lockstep model. Engine policy is one 80 ms frame per resumable Stage 0 append with greedy one-token decoding; session policy is the bundled voice prompt and the persona text, resolved once at open into the server-owned runtime configuration the worker replays as its prefill.

PRIVATE_RUNTIME_CONFIG_KEYS module-attribute 

PRIVATE_RUNTIME_CONFIG_KEYS = frozenset(
    {
        "personaplex_prefill_slots",
        "personaplex_model_path",
        "personaplex_voice_prompt",
        "personaplex_persona",
    }
)

PersonaPlexDuplexPlugin

Bases: DuplexModelPlugin

PersonaPlex-owned sampling policy, append planning, session state and output projection.

data_plane instance-attribute 

data_plane = PersonaPlexDataPlaneSession(encode_audio)

plugin_id class-attribute instance-attribute 

plugin_id = 'personaplex'

private_runtime_config_keys class-attribute instance-attribute 

private_runtime_config_keys = PRIVATE_RUNTIME_CONFIG_KEYS

silence_continuation_sample_rate_hz class-attribute instance-attribute 

silence_continuation_sample_rate_hz = SAMPLE_RATE

silence_continuation_samples class-attribute instance-attribute 

silence_continuation_samples = FRAME_SIZE

capabilities 

capabilities(*, max_sessions: int) -> DuplexCapabilities

configure_sampling_params 

configure_sampling_params(
    *,
    runtime_config: dict[str, object],
    defaults: tuple[object, ...],
) -> tuple[object, ...]

create_session_state 

create_session_state() -> PersonaPlexSessionState

decide_output 

decide_output(
    *,
    stage_id: int,
    final_stage_id: int,
    segment_finished: bool,
    segment_token_ids: tuple[int, ...],
    segment_output_metadata: dict[str, object],
    output: object,
) -> DuplexOutputDecision | None

plan_append 

plan_append(
    *,
    request_id: str,
    fence: DuplexFence,
    session_config: dict[str, object],
    runtime_config: dict[str, object],
    seq: int,
    turn_seq: int,
    payload: object,
    final: bool,
    sampling_params: object,
) -> DuplexAppendPlan

prepare_runtime_config async 

prepare_runtime_config(
    config: DuplexSessionConfig,
    *,
    model_config: ModelConfig | None,
) -> dict[str, object]

runtime_config_for_update 

runtime_config_for_update(
    config: DuplexSessionConfig,
    current: Mapping[str, object],
) -> dict[str, object]

PersonaPlexSessionState dataclass

Bases: DefaultDuplexModelSessionState

Per-session model state: the framework flags plus the 80 ms framing buffer.

audio_buffer class-attribute instance-attribute 

audio_buffer: PersonaPlexPcmAppendBuffer = field(
    default_factory=PersonaPlexPcmAppendBuffer
)

persona_text 

persona_text(value: object) -> str

prefill_slot_count 

prefill_slot_count(
    runtime_config: Mapping[str, object],
) -> int

voice_name 

voice_name(value: object) -> str

A bundled .pt basename (NATF2.pt); paths are refused so the worker only reads the checkpoint.