vllm.models.muse_spark.nvidia.multimodal

Muse Spark Omni for NVIDIA GPUs: the shared Omni towers over the CUDA text stack.

Classes:

• MuseSparkForConditionalGeneration –

  Muse Spark Omni: text stack plus NaViT vision and waveform audio.

MuseSparkForConditionalGeneration

Bases: MuseSparkOmniMixin, MuseSparkForCausalLM, SupportsMultiModal

Muse Spark Omni: text stack plus NaViT vision and waveform audio.

Source code in vllm/models/muse_spark/nvidia/multimodal.py

@register_omni_processor
class MuseSparkForConditionalGeneration(
    MuseSparkOmniMixin, MuseSparkForCausalLM, SupportsMultiModal
):
    """Muse Spark Omni: text stack plus NaViT vision and waveform audio."""

    hf_to_vllm_mapper = MuseSparkForCausalLM.hf_to_vllm_mapper | OMNI_WEIGHTS_MAPPER