Update app.py

app.py

@@ -193,13 +193,19 @@ def generate_image(model_name: str, text: str, image: Image.Image,
 
     images = [image.convert("RGB")]
 
-    messages = [
-        {
-            "role": "user",
-            "content": [{"type": "image"}] + [{"type": "text", "text": text}]
-        }
-    ]
-    prompt = processor.apply_chat_template(messages, add_generation_prompt=True)
+    # Create the prompt based on the specific model's requirements
+    if model_name == "PaddleOCR":
+        # PaddleOCR's template expects a single string with an image placeholder
+        messages = [
+            {"role": "user", "content": f"<image>\n{text}"}
+        ]
+    else:
+        # Standard format for Nanonets and Dots.OCR
+        messages = [
+            {"role": "user", "content": [{"type": "image"}] + [{"type": "text", "text": text}]}
+        ]
+
+    prompt = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
     inputs = processor(text=prompt, images=images, return_tensors="pt").to(device)
 
     streamer = TextIteratorStreamer(processor, skip_prompt=True, skip_special_tokens=True)