Update app.py

app.py

@@ -1,56 +1,37 @@
-import torch
+# app.py — use Qwen2-VL via together.ai or openrouter (free tier)
 import gradio as gr
-from transformers import AutoProcessor, AutoModelForVision2Seq
+import base64
+from io import BytesIO
 from PIL import Image
+from openai import OpenAI
 
-model_id = "HuggingFaceTB/SmolVLM-Instruct"
-
-print("Loading model...")
-processor = AutoProcessor.from_pretrained(model_id)
-model = AutoModelForVision2Seq.from_pretrained(
-    model_id,
-    torch_dtype=torch.float32,
-    device_map="auto"
+client = OpenAI(
+    api_key="your_free_api_key",  # openrouter.ai free tier
+    base_url="https://openrouter.ai/api/v1"
 )
-model.eval()
-print("Model ready!")
 
 def extract_text(image: Image.Image) -> str:
-    messages = [
-        {
+    buffered = BytesIO()
+    image.save(buffered, format="PNG")
+    img_b64 = base64.b64encode(buffered.getvalue()).decode()
+
+    response = client.chat.completions.create(
+        model="qwen/qwen2-vl-7b-instruct:free",  # free model
+        messages=[{
             "role": "user",
             "content": [
-                {"type": "image"},
-                {"type": "text", "text": (
-                    "Extract all text from this image exactly as it appears. "
-                    "Preserve question numbers, options A B C D, tables, "
-                    "and any mathematical or chemical expressions. "
-                    "Format clearly."
-                )}
+                {"type": "image_url", "image_url": {"url": f"data:image/png;base64,{img_b64}"}},
+                {"type": "text", "text": "Extract all text exactly as it appears. Preserve question numbers, options A B C D, tables, and equations."}
             ]
-        }
-    ]
-
-    prompt = processor.apply_chat_template(messages, add_generation_prompt=True)
-    inputs = processor(text=prompt, images=[image], return_tensors="pt")
-
-    with torch.no_grad():
-        outputs = model.generate(
-            **inputs,
-            max_new_tokens=1024,
-            do_sample=False
-        )
-
-    generated = outputs[0][inputs["input_ids"].shape[1]:]
-    return processor.decode(generated, skip_special_tokens=True)
-
+        }]
+    )
+    return response.choices[0].message.content
 
 demo = gr.Interface(
     fn=extract_text,
     inputs=gr.Image(type="pil", label="Upload NEET Question Image"),
     outputs=gr.Textbox(label="Extracted Text", lines=20),
-    title="NEET Question Extractor",
-    description="Upload a scanned NEET question paper image to extract text"
+    title="NEET Question Extractor"
 )
 
 demo.launch()