Add vLLM-Omni as a self-hosting option for Higgs Audio v3
#16
by linyueqian - opened
Adds a vLLM-Omni Usage subsection alongside SGLang Usage. vLLM-Omni has first-class Higgs Audio v3 TTS support merged upstream (vllm-project/vllm-omni#4169, perf #4204), serving the same OpenAI-compatible /v1/audio/speech API with zero-shot voice cloning. Serve command mirrors the official vLLM-Omni recipe.
popsoda2002 changed pull request status to merged
Hi @linyueqian Thanks for the contribution, merged 😀