diff --git a/docs/getting_started/installation/gpu/cuda.inc.md b/docs/getting_started/installation/gpu/cuda.inc.md
index 45fa69ddb..932b54f0d 100644
--- a/docs/getting_started/installation/gpu/cuda.inc.md
+++ b/docs/getting_started/installation/gpu/cuda.inc.md
@@ -29,6 +29,11 @@ uv pip install vllm --torch-backend=auto
 uv pip install vllm-omni
 ```
 
+To run Gradio demos, also install the optional extras:
+```bash
+uv pip install 'vllm-omni[demo]'
+```
+
 # --8<-- [end:pre-built-wheels]
 
 # --8<-- [start:build-wheel-from-source]
@@ -50,6 +55,11 @@ cd vllm-omni
 uv pip install -e .
 ```
 
+To run Gradio demos, install with optional extras:
+```bash
+uv pip install -e '.[demo]'
+```
+
 <details><summary>(Optional) Installation of vLLM from source</summary>
 If you want to check, modify or debug with source code of vLLM, install the library from source with the following instructions:
 
diff --git a/docs/user_guide/examples/offline_inference/text_to_image.md b/docs/user_guide/examples/offline_inference/text_to_image.md
index cea136b35..3a97ffbf7 100644
--- a/docs/user_guide/examples/offline_inference/text_to_image.md
+++ b/docs/user_guide/examples/offline_inference/text_to_image.md
@@ -256,6 +256,15 @@ lora_adapter/
 
 ## Web UI Demo
 
+!!! note "Gradio is an optional dependency"
+    The Gradio demo requires the `[demo]` extras. Install them first:
+
+    ```bash
+    pip install 'vllm-omni[demo]'
+    ```
+
+    Or, if installing from source: `pip install -e '.[demo]'`
+
 Launch the Gradio demo:
 
 ```bash
diff --git a/docs/user_guide/examples/online_serving/fish_speech.md b/docs/user_guide/examples/online_serving/fish_speech.md
index 2c76b1bca..7322d06aa 100644
--- a/docs/user_guide/examples/online_serving/fish_speech.md
+++ b/docs/user_guide/examples/online_serving/fish_speech.md
@@ -11,6 +11,15 @@ Source <https://github.com/vllm-project/vllm-omni/tree/main/examples/online_serv
 
 ## Gradio Demo
 
+!!! note "Gradio is an optional dependency"
+    The Gradio demo requires the `[demo]` extras. Install them first:
+
+    ```bash
+    pip install 'vllm-omni[demo]'
+    ```
+
+    Or, if installing from source: `pip install -e '.[demo]'`
+
 An interactive Gradio demo is available with text-to-speech synthesis, voice cloning, and streaming support.
 
 ```bash
diff --git a/docs/user_guide/examples/online_serving/image_to_image.md b/docs/user_guide/examples/online_serving/image_to_image.md
index b19e9462d..f40173c4f 100644
--- a/docs/user_guide/examples/online_serving/image_to_image.md
+++ b/docs/user_guide/examples/online_serving/image_to_image.md
@@ -122,6 +122,15 @@ python openai_chat_client.py --input input1.png input2.png --prompt "Combine the
 
 ### Method 4: Using Gradio Demo
 
+!!! note "Gradio is an optional dependency"
+    The Gradio demo requires the `[demo]` extras. Install them first:
+
+    ```bash
+    pip install 'vllm-omni[demo]'
+    ```
+
+    Or, if installing from source: `pip install -e '.[demo]'`
+
 ```bash
 python gradio_demo.py
 # Visit http://localhost:7861
diff --git a/docs/user_guide/examples/online_serving/qwen2_5_omni.md b/docs/user_guide/examples/online_serving/qwen2_5_omni.md
index 084e6a3f0..435764692 100644
--- a/docs/user_guide/examples/online_serving/qwen2_5_omni.md
+++ b/docs/user_guide/examples/online_serving/qwen2_5_omni.md
@@ -149,6 +149,15 @@ python openai_chat_completion_client_for_multimodal_generation.py \
 
 ## Run Local Web UI Demo
 
+!!! note "Gradio is an optional dependency"
+    The Gradio demo requires the `[demo]` extras. Install them first:
+
+    ```bash
+    pip install 'vllm-omni[demo]'
+    ```
+
+    Or, if installing from source: `pip install -e '.[demo]'`
+
 This Web UI demo allows users to interact with the model through a web browser.
 
 ### Running Gradio Demo
diff --git a/docs/user_guide/examples/online_serving/qwen3_omni.md b/docs/user_guide/examples/online_serving/qwen3_omni.md
index e44565a21..69de24852 100644
--- a/docs/user_guide/examples/online_serving/qwen3_omni.md
+++ b/docs/user_guide/examples/online_serving/qwen3_omni.md
@@ -163,6 +163,15 @@ python openai_chat_completion_client_for_multimodal_generation.py \
 
 ## Run Local Web UI Demo
 
+!!! note "Gradio is an optional dependency"
+    The Gradio demo requires the `[demo]` extras. Install them first:
+
+    ```bash
+    pip install 'vllm-omni[demo]'
+    ```
+
+    Or, if installing from source: `pip install -e '.[demo]'`
+
 This Web UI demo allows users to interact with the model through a web browser.
 
 ### Running Gradio Demo
diff --git a/docs/user_guide/examples/online_serving/qwen3_tts.md b/docs/user_guide/examples/online_serving/qwen3_tts.md
index 03135085d..401a5c2e9 100644
--- a/docs/user_guide/examples/online_serving/qwen3_tts.md
+++ b/docs/user_guide/examples/online_serving/qwen3_tts.md
@@ -19,6 +19,15 @@ Please refer to [README.md](https://github.com/vllm-project/vllm-omni/tree/main/
 
 ## Gradio Demo
 
+!!! note "Gradio is an optional dependency"
+    The Gradio demo requires the `[demo]` extras. Install them first:
+
+    ```bash
+    pip install 'vllm-omni[demo]'
+    ```
+
+    Or, if installing from source: `pip install -e '.[demo]'`
+
 Two interactive Gradio demos are available, both supporting all 3 task types:
 
 | Demo     | File                     | Transport    | Streaming Quality                                  |
diff --git a/docs/user_guide/examples/online_serving/text_to_image.md b/docs/user_guide/examples/online_serving/text_to_image.md
index 2e79749b3..69c1480e3 100644
--- a/docs/user_guide/examples/online_serving/text_to_image.md
+++ b/docs/user_guide/examples/online_serving/text_to_image.md
@@ -105,6 +105,15 @@ python openai_chat_client.py --prompt "A beautiful landscape painting" --output
 
 ### Method 4: Using Gradio Demo
 
+!!! note "Gradio is an optional dependency"
+    The Gradio demo requires the `[demo]` extras. Install them first:
+
+    ```bash
+    pip install 'vllm-omni[demo]'
+    ```
+
+    Or, if installing from source: `pip install -e '.[demo]'`
+
 ```bash
 python gradio_demo.py
 # Visit http://localhost:7860
diff --git a/examples/offline_inference/text_to_image/gradio_demo.py b/examples/offline_inference/text_to_image/gradio_demo.py
index 928325cd6..09e011240 100644
--- a/examples/offline_inference/text_to_image/gradio_demo.py
+++ b/examples/offline_inference/text_to_image/gradio_demo.py
@@ -1,7 +1,10 @@
 import argparse
 from functools import lru_cache
 
-import gradio as gr
+try:
+    import gradio as gr
+except ImportError:
+    raise ImportError("gradio is required to run this demo. Install it with: pip install 'vllm-omni[demo]'") from None
 import torch
 
 from vllm_omni.entrypoints.omni import Omni
diff --git a/examples/online_serving/fish_speech/gradio_demo.py b/examples/online_serving/fish_speech/gradio_demo.py
index a8e51b274..7db4e0441 100644
--- a/examples/online_serving/fish_speech/gradio_demo.py
+++ b/examples/online_serving/fish_speech/gradio_demo.py
@@ -16,7 +16,10 @@ import argparse
 import base64
 import io
 
-import gradio as gr
+try:
+    import gradio as gr
+except ImportError:
+    raise ImportError("gradio is required to run this demo. Install it with: pip install 'vllm-omni[demo]'") from None
 import httpx
 import numpy as np
 import soundfile as sf
diff --git a/examples/online_serving/image_to_image/gradio_demo.py b/examples/online_serving/image_to_image/gradio_demo.py
index 8cad48279..7cf062ca8 100644
--- a/examples/online_serving/image_to_image/gradio_demo.py
+++ b/examples/online_serving/image_to_image/gradio_demo.py
@@ -10,7 +10,10 @@ import argparse
 import base64
 from io import BytesIO
 
-import gradio as gr
+try:
+    import gradio as gr
+except ImportError:
+    raise ImportError("gradio is required to run this demo. Install it with: pip install 'vllm-omni[demo]'") from None
 import requests
 from PIL import Image
 
diff --git a/examples/online_serving/qwen2_5_omni/gradio_demo.py b/examples/online_serving/qwen2_5_omni/gradio_demo.py
index 4fdef5248..11dd76088 100644
--- a/examples/online_serving/qwen2_5_omni/gradio_demo.py
+++ b/examples/online_serving/qwen2_5_omni/gradio_demo.py
@@ -6,7 +6,10 @@ import random
 from pathlib import Path
 from typing import Any
 
-import gradio as gr
+try:
+    import gradio as gr
+except ImportError:
+    raise ImportError("gradio is required to run this demo. Install it with: pip install 'vllm-omni[demo]'") from None
 import numpy as np
 import soundfile as sf
 import torch
diff --git a/examples/online_serving/qwen3_omni/gradio_demo.py b/examples/online_serving/qwen3_omni/gradio_demo.py
index 76c4e311a..e0516dc4a 100644
--- a/examples/online_serving/qwen3_omni/gradio_demo.py
+++ b/examples/online_serving/qwen3_omni/gradio_demo.py
@@ -6,7 +6,10 @@ import random
 from pathlib import Path
 from typing import Any
 
-import gradio as gr
+try:
+    import gradio as gr
+except ImportError:
+    raise ImportError("gradio is required to run this demo. Install it with: pip install 'vllm-omni[demo]'") from None
 import numpy as np
 import soundfile as sf
 import torch
diff --git a/examples/online_serving/qwen3_tts/gradio_demo.py b/examples/online_serving/qwen3_tts/gradio_demo.py
index ccf56d815..b263cb8a2 100644
--- a/examples/online_serving/qwen3_tts/gradio_demo.py
+++ b/examples/online_serving/qwen3_tts/gradio_demo.py
@@ -24,7 +24,10 @@ import io
 import json
 import logging
 
-import gradio as gr
+try:
+    import gradio as gr
+except ImportError:
+    raise ImportError("gradio is required to run this demo. Install it with: pip install 'vllm-omni[demo]'") from None
 import httpx
 import numpy as np
 import soundfile as sf
diff --git a/examples/online_serving/qwen3_tts/tts_common.py b/examples/online_serving/qwen3_tts/tts_common.py
index 4a44b997f..20b78517e 100644
--- a/examples/online_serving/qwen3_tts/tts_common.py
+++ b/examples/online_serving/qwen3_tts/tts_common.py
@@ -3,7 +3,10 @@
 import base64
 import io
 
-import gradio as gr
+try:
+    import gradio as gr
+except ImportError:
+    raise ImportError("gradio is required to run this demo. Install it with: pip install 'vllm-omni[demo]'") from None
 import httpx
 import numpy as np
 import soundfile as sf
diff --git a/examples/online_serving/text_to_image/gradio_demo.py b/examples/online_serving/text_to_image/gradio_demo.py
index 608db9a23..c8329c02a 100644
--- a/examples/online_serving/text_to_image/gradio_demo.py
+++ b/examples/online_serving/text_to_image/gradio_demo.py
@@ -10,7 +10,10 @@ import argparse
 import base64
 from io import BytesIO
 
-import gradio as gr
+try:
+    import gradio as gr
+except ImportError:
+    raise ImportError("gradio is required to run this demo. Install it with: pip install 'vllm-omni[demo]'") from None
 import requests
 from PIL import Image
 
diff --git a/examples/online_serving/voxtral_tts/gradio_demo.py b/examples/online_serving/voxtral_tts/gradio_demo.py
index dd987f229..35d6b590c 100644
--- a/examples/online_serving/voxtral_tts/gradio_demo.py
+++ b/examples/online_serving/voxtral_tts/gradio_demo.py
@@ -20,7 +20,10 @@ from datetime import datetime
 from pathlib import Path
 from typing import Any
 
-import gradio as gr
+try:
+    import gradio as gr
+except ImportError:
+    raise ImportError("gradio is required to run this demo. Install it with: pip install 'vllm-omni[demo]'") from None
 import httpx
 import numpy as np
 import soundfile as sf
diff --git a/pyproject.toml b/pyproject.toml
index aedcaf132..4e87e0f30 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -56,6 +56,10 @@ dev = [
     "mistune>=3.2.0", # for example tests
 ]
 
+demo = [
+    "gradio>=6.7.0",
+]
+
 docs = [
     "mkdocs>=1.5.0",
     "mkdocs-api-autonav",
diff --git a/requirements/common.txt b/requirements/common.txt
index 1b65974fe..138a61ed2 100644
--- a/requirements/common.txt
+++ b/requirements/common.txt
@@ -4,7 +4,6 @@ librosa>=0.11.0
 resampy>=0.4.3
 diffusers>=0.36.0
 accelerate==1.12.0
-gradio==5.50
 soundfile>=0.13.1
 cache-dit==1.3.0
 tqdm>=4.66.0
@@ -18,3 +17,4 @@ prettytable>=3.8.0
 aenum==3.1.16
 pyzmq>=25.0.0
 janus>=1.0.0
+pydub