requirements.txt

transformers
torch
gradio
accelerate
pillow

app.py

@@ -0,0 +1,28 @@
+import gradio as gr
+from transformers import AutoProcessor, AutoModelForVision2Seq
+from PIL import Image
+
+# Load model and processor
+processor = AutoProcessor.from_pretrained("Qwen/Qwen2.5-VL-7B-Instruct")
+model = AutoModelForVision2Seq.from_pretrained("Qwen/Qwen2.5-VL-7B-Instruct", device_map="auto")
+
+# Prompt specifically designed for CD spines
+prompt = "What is the album title and artist on this CD spine?"
+
+# Inference function
+def extract_text(image: Image.Image):
+    inputs = processor(prompt=prompt, images=image, return_tensors="pt").to(model.device)
+    generated_ids = model.generate(**inputs, max_new_tokens=128)
+    generated_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
+    return {"text": generated_text}
+
+# Gradio interface
+iface = gr.Interface(
+    fn=extract_text,
+    inputs="image",
+    outputs="json",
+    title="Qwen2.5-VL: CD Spine Reader",
+    description="Upload an image of a CD spine to get the album title and artist using Qwen2.5-VL-7B-Instruct."
+)
+
+iface.launch()