From 40a000de059886fb799b013fb7bf2ebafc8ab978 Mon Sep 17 00:00:00 2001
From: Sy03 <1370724210@qq.com>
Date: Wed, 16 Sep 2026 16:00:45 +0800
Subject: [PATCH] Document multi-API serving tradeoffs in TTS profile

Signed-off-by: Sy03 <1370724210@qq.com>
---
 vllm_omni/deploy/qwen3_tts_high_concurrency.yaml | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/vllm_omni/deploy/qwen3_tts_high_concurrency.yaml b/vllm_omni/deploy/qwen3_tts_high_concurrency.yaml
index 7a73dd50a..0526fb864 100644
--- a/vllm_omni/deploy/qwen3_tts_high_concurrency.yaml
+++ b/vllm_omni/deploy/qwen3_tts_high_concurrency.yaml
@@ -9,6 +9,15 @@
 # Profile validated for the c64 / PROMPTS=512 performance experiments:
 # Stage 0 talker on GPU 0 with S0=64, Stage 1 Code2Wav on GPU 1 with S1=10.
 #
+# For CPU-bound frontend or orchestration workloads, this profile can be used
+# with ``--api-server-count 2`` or higher to run multiple API frontends over
+# one shared set of local stage engines. This can improve request parsing,
+# multimodal/audio preprocessing, and response handling under concurrency;
+# it does not duplicate GPU engines or make a single GPU-bound request faster.
+# Each frontend keeps process-local state, so runtime voice mutations, sleep /
+# wakeup, async video jobs, and resumable duplex sessions are not supported
+# with more than one API worker. See docs/cli/serve.md for the full boundary.
+#
 async_chunk: true
 
 connectors: