Skip to content

vllm_omni.model_executor.models.personaplex.duplex.config

Configuration for the PersonaPlex full-duplex backend.

PersonaPlex (nvidia/personaplex-7b-v1) is a Moshi finetune: a pure-lockstep full-duplex speech-to-speech model running at the Mimi codec frame rate (12.5 Hz / 80 ms). One config drives both the offline driver and the duplex adapter. Defaults mirror the PersonaPlex reference loop.

DEFAULT_PERSONA module-attribute 

DEFAULT_PERSONA = "You are a wise and friendly teacher. Answer questions or provide advice in a clear and engaging way."

FRAME_RATE module-attribute 

FRAME_RATE = 12.5

FRAME_SIZE module-attribute 

FRAME_SIZE = int(SAMPLE_RATE / FRAME_RATE)

SAMPLE_RATE module-attribute 

SAMPLE_RATE = 24000

PersonaPlexConfig dataclass

Immutable session configuration for a PersonaPlex conversation.

Attributes:

Name Type Description
hf_repo str

HuggingFace repo holding the weights, Mimi codec and tokenizer.
voice_prompt str

Voice-clone reference. Either a bundled basename ("NATF2.pt" / "NATM1.pt" from voices.tgz) or a path to a .pt embedding bundle or a reference .wav.
persona str

System role text; injected as <system> ... <system> into the inner-monologue stream at session start.
device str

Torch device for the backend ("cuda" / "cpu").
cpu_offload bool

Offload LM layers to CPU when GPU memory is tight (needs accelerate).
batch_size int

Concurrent conversation slots sharing one engine. 1 is the single-session path; > 1 enables elastic batching with per-slot recycle for new callers.

Note: the native stepper decodes greedily (argmax) for both the text head and the depformer, so there are no sampling knobs here yet. Temperature / top-k / seed fields will be added if and when a sampling path is wired in.

batch_size class-attribute instance-attribute 

batch_size: int = 1

cpu_offload class-attribute instance-attribute 

cpu_offload: bool = False

device class-attribute instance-attribute 

device: str = 'cuda'

frame_size property 

frame_size: int

hf_repo class-attribute instance-attribute 

hf_repo: str = 'nvidia/personaplex-7b-v1'

persona class-attribute instance-attribute 

persona: str = DEFAULT_PERSONA

sample_rate property 

sample_rate: int

voice_prompt class-attribute instance-attribute 

voice_prompt: str = 'NATF2.pt'