Update app.py

app.py

@@ -5,7 +5,7 @@ from typing import Iterator, Callable
 
 import gradio as gr
 import spaces
-from transformers import Qwen2_5_VLForConditionalGeneration, AutoProcessor
+from transformers import Qwen2_5_VLForConditionalGeneration, Qwen3VLMoeForConditionalGeneration, AutoProcessor
 from qwen_vl_utils import process_vision_info
 
 subprocess.run(
@@ -19,6 +19,8 @@ MODEL_NAMES = {
     "Qwen2.5-VL-7B-Instruct": "Qwen/Qwen2.5-VL-7B-Instruct",
     "Qwen2.5-VL-3B-Instruct": "Qwen/Qwen2.5-VL-3B-Instruct",
     "Qwen2.5-VL-32B-Instruct": "Qwen/Qwen2.5-VL-32B-Instruct",
+    "Qwen3-VL-4B-Instruct": "Qwen/Qwen3-VL-4B-Instruct",
+    "Qwen3-VL-8B-Instruct": "Qwen/Qwen3-VL-8B-Instruct",
 }
 
 
@@ -43,11 +45,18 @@ def run_inference(model_key, input_type, text, image, video, fps, system_prompt,
     Load the selected Qwen2.5-VL model and run inference on text, image, or video.
     """
     model_id = MODEL_NAMES[model_key]
-    model = Qwen2_5_VLForConditionalGeneration.from_pretrained(
-        model_id,
-        torch_dtype="auto",
-        device_map="auto",
-    )
+    if "Qwen3" in model_id:
+        model = Qwen3VLMoeForConditionalGeneration.from_pretrained(
+            model_id,
+            torch_dtype="auto",
+            device_map="auto",
+        )
+    else:
+        model = Qwen2_5_VLForConditionalGeneration.from_pretrained(
+            model_id,
+            torch_dtype="auto",
+            device_map="auto",
+        )
     processor = AutoProcessor.from_pretrained(model_id)
 
     # Text-only inference