vllm_omni.model_executor.models.ming_tts.pipeline ¶

MING_TTS_MOE_PIPELINE = PipelineConfig(
    model_type="ming_tts_moe",
    model_arch="MingTTSForConditionalGeneration",
    hf_architectures=(
        "MingTTSForConditionalGeneration",
        "BailingMMNativeForConditionalGeneration",
    ),
    hf_config_predicate=lambda c: (
        str(getattr(c, "model_type", "")) == "bailingmm"
    ),
    stages=MING_TTS_PIPELINE.stages,
)

MING_TTS_PIPELINE = PipelineConfig(
    model_type="ming_tts",
    model_arch="MingTTSForConditionalGeneration",
    hf_architectures=(
        "MingTTSForConditionalGeneration",
        "BailingMMNativeForConditionalGeneration",
    ),
    hf_config_predicate=lambda c: (
        str(getattr(c, "model_type", "")) == "dense"
    ),
    stages=(
        StagePipelineConfig(
            stage_id=0,
            model_stage="llm",
            execution_type=StageExecutionType.LLM_AR,
            input_sources=(),
            owns_tokenizer=True,
            hf_config_name="llm_config",
            engine_output_type="latent",
            async_chunk_process_next_stage_input_func=f"{_PROC}.llm2audio_vae_async_chunk",
            sampling_constraints={
                "temperature": 0.0,
                "top_p": 1.0,
                "top_k": -1,
                "max_tokens": 512,
                "detokenize": True,
            },
        ),
        StagePipelineConfig(
            stage_id=1,
            model_stage="audio_vae",
            execution_type=StageExecutionType.LLM_GENERATION,
            input_sources=(0,),
            final_output=True,
            final_output_type="audio",
            hf_config_name="llm_config",
            engine_output_type="audio",
            sync_process_input_func=f"{_PROC}.llm2audio_vae",
            sampling_constraints={
                "temperature": 0.0,
                "top_p": 1.0,
                "top_k": -1,
                "max_tokens": 1,
                "detokenize": False,
            },
        ),
    ),
)