Update space

app.py

@@ -1,64 +1,91 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+import base64
+import requests
+from PIL import Image
+import io
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+API_KEY = "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"
+IMAGE_MODEL = "OpenGVLab/InternVL3-14B"
 
+def extract_passport_info(images, document_type):
+    results = []
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+    for image in images:
+        # Convert image to base64
+        buffered = io.BytesIO()
+        image.save(buffered, format="JPEG")
+        encoded_image = base64.b64encode(buffered.getvalue()).decode("utf-8")
+        data_url = f"data:image/jpeg;base64,{encoded_image}"
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+        # Prompt to extract full passport data
+        prompt = (
+            f"Extract all passport information from the uploaded {document_type} image. "
+            "Include MRZ (if present), full name, passport number, nationality, gender, "
+            "date of birth, date of issue, expiry date, issuing country, and any other text or labels in other languages. "
+            "Return the result in a JSON format."
+        )
 
-    messages.append({"role": "user", "content": message})
+        # OpenRouter Payload
+        payload = {
+            "model": IMAGE_MODEL,
+            "messages": [
+                {
+                    "role": "user",
+                    "content": [
+                        {"type": "text", "text": prompt},
+                        {"type": "image_url", "image_url": {"url": data_url}},
+                    ],
+                }
+            ],
+        }
 
-    response = ""
+        headers = {
+            "Authorization": f"Bearer {API_KEY}",
+            "Content-Type": "application/json"
+        }
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+        try:
+            response = requests.post("https://openrouter.ai/api/v1/chat/completions", headers=headers, json=payload)
+            result = response.json()
+            print("📡 Status:", response.status_code)
+            print("📡 Raw Result:", result)
 
-        response += token
-        yield response
+            if "choices" in result:
+                extracted = result["choices"][0]["message"]["content"]
+                results.append({
+                    "document_type": document_type,
+                    "extracted_info": extracted
+                })
+            else:
+                results.append({
+                    "document_type": document_type,
+                    "extracted_info": "❌ No data extracted"
+                })
 
+        except Exception as e:
+            results.append({
+                "document_type": document_type,
+                "extracted_info": f"⚠️ Error: {str(e)}"
+            })
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
+    return results
+
+
+# Gradio UI
+demo = gr.Interface(
+    fn=extract_passport_info,
+    inputs=[
+        gr.Image(type="pil", label="Upload Passport/Document Images", multiple=True),
+        gr.Dropdown(
+            choices=["passport_front", "passport_back", "photo", "hotel_reservation"],
+            label="Document Type",
+            value="passport_front",
+        )
     ],
+    outputs="json",
+    title="Passport & Document Info Extractor",
+    description="Upload one or more document images. Extracted information will include MRZ and all available text, structured in JSON format.",
 )
 
-
 if __name__ == "__main__":
     demo.launch()

requirements.txt

@@ -1 +1,4 @@
-huggingface_hub==0.25.2
+gradio==3.50.2
+requests
+python-dotenv
+Pillow