app.py

from transformers import BlipProcessor, BlipForConditionalGeneration
import gradio as gr

# Load processor and model
processor = BlipProcessor.from_pretrained(
    "Salesforce/blip-image-captioning-base"
)

model = BlipForConditionalGeneration.from_pretrained(
    "Salesforce/blip-image-captioning-base"
)

# Function
def generate_caption(image):

    inputs = processor(images=image, return_tensors="pt")

    output = model.generate(**inputs)

    caption = processor.decode(
        output[0],
        skip_special_tokens=True
    )

    return caption

# Gradio interface
demo = gr.Interface(
    fn=generate_caption,
    inputs=gr.Image(type="pil"),
    outputs="text",
    title="AI Image Caption Generator",
    description="Upload image and AI describes it."
)

# Launch app
demo.launch()