Update app.py

app.py

@@ -32,6 +32,15 @@ model_m = Qwen2_5_VLForConditionalGeneration.from_pretrained(
     torch_dtype=torch.float16
 ).to(device).eval()
 
+# Load VLM-R1-Qwen2.5VL-3B-OVD-0321
+MODEL_ID_X = "omlab/VLM-R1-Qwen2.5VL-3B-OVD-0321"
+processor_x = AutoProcessor.from_pretrained(MODEL_ID_X, trust_remote_code=True)
+model_x = Qwen2_5_VLForConditionalGeneration.from_pretrained(
+    MODEL_ID_X,
+    trust_remote_code=True,
+    torch_dtype=torch.float16
+).to(device).eval()
+
 # Load Megalodon-OCR-Sync-0713
 MODEL_ID_T = "prithivMLmods/Megalodon-OCR-Sync-0713"
 processor_t = AutoProcessor.from_pretrained(MODEL_ID_T, trust_remote_code=True)
@@ -101,6 +110,9 @@ def generate_image(model_name: str, text: str, image: Image.Image,
     elif model_name == "DeepEyes-7B-Thinking":
         processor = processor_y
         model = model_y
+    elif model_name == "VLM-R1-Qwen2.5VL":
+        processor = processor_x
+        model = model_x
     else:
         yield "Invalid model selected.", "Invalid model selected."
         return
@@ -157,6 +169,9 @@ def generate_video(model_name: str, text: str, video_path: str,
     elif model_name == "DeepEyes-7B-Thinking":
         processor = processor_y
         model = model_y
+    elif model_name == "VLM-R1-Qwen2.5VL":
+        processor = processor_x
+        model = model_x
     else:
         yield "Invalid model selected.", "Invalid model selected."
         return
@@ -272,7 +287,7 @@ with gr.Blocks(css=css, theme="bethecloud/storj_theme") as demo:
                     markdown_output = gr.Markdown(label="(Result.md)")
 
             model_choice = gr.Radio(
-                choices=["Camel-Doc-OCR-062825", "GLM-4.1V-9B-Thinking", "Megalodon-OCR-Sync-0713", "DeepEyes-7B-Thinking"],
+                choices=["Camel-Doc-OCR-062825", "GLM-4.1V-9B-Thinking", "Megalodon-OCR-Sync-0713", "DeepEyes-7B-Thinking", "VLM-R1-Qwen2.5VL"],
                 label="Select Model",
                 value="Camel-Doc-OCR-062825"
             )