vllm_omni.model_executor.models.step_audio2 ¶
Modules:
| Name | Description |
|---|---|
pipeline | Step-Audio2 pipeline topologies (frozen). |
step_audio2 | |
step_audio2_constants | Step-Audio2 configuration constants - Single Source of Truth. |
step_audio2_dit_trt | TensorRT acceleration for the Step-Audio2 / CosyVoice2 flow DiT estimator. |
step_audio2_thinker | Step-Audio2 Thinker - Stage 1 LLM for Audio Understanding |
step_audio2_token2wav | |