From 0aa72b952f29adaecdbe60a40966b761270f41e7 Mon Sep 17 00:00:00 2001
From: Sihao Li <111170255+GG-li@users.noreply.github.com>
Date: Wed, 21 Jan 2026 17:44:17 +0800
Subject: [PATCH] [bugfix] support text + audio mixed output (#843)

Signed-off-by: GG-li <3226868735@qq.com>
---
 vllm_omni/entrypoints/openai/serving_chat.py | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/vllm_omni/entrypoints/openai/serving_chat.py b/vllm_omni/entrypoints/openai/serving_chat.py
index 6320cb73a..8a62fd16c 100644
--- a/vllm_omni/entrypoints/openai/serving_chat.py
+++ b/vllm_omni/entrypoints/openai/serving_chat.py
@@ -1324,11 +1324,21 @@ class OmniOpenAIServingChat(OpenAIServingChat, AudioMixin):
         prompt_token_ids = None
         kv_transfer_params = None
 
+        # Build requested modalities set for filtering
+        requested_modalities = (
+            set(request.modalities) if hasattr(request, "modalities") and request.modalities else None
+        )
+
         for omni_outputs in final_outputs:
             choices_data = []
             if omni_outputs.request_output is not None and not getattr(omni_outputs.request_output, "finished", False):
                 continue
 
+            # Filter outputs based on requested modalites
+            if requested_modalities is not None and omni_outputs.final_output_type not in requested_modalities:
+                logger.warning(f"final output type: {omni_outputs.final_output_type} is not needed by the request")
+                continue
+
             if omni_outputs.final_output_type == "text":
                 (
                     choices_data,