import streamlit as st import requests import os import base64 from PIL import Image from io import BytesIO # Correctly import BytesIO from the io module # Set page title and layout st.set_page_config(page_title="Image Caption Generator", layout="wide") # API key from environment variable API_KEY = os.environ.get("NEBIUS_API_KEY") if not API_KEY: st.error("API key not found. Please set the `NEBIUS_API_KEY` environment variable.") # Function to call Nebius API def generate_caption(image_base64, api_key): api_url = "https://api.studio.nebius.ai/v1/chat/completions" headers = {"Authorization": f"Bearer {api_key}"} payload = { "model": "Qwen/Qwen2-VL-7B-Instruct", "messages": [ { "role": "system", "content": """You are an image to text prompt converter for AI image generation""", }, { "role": "user", "content": [ {"type": "text", "text": "Describe This Image In Detail under 100 words in this format: [image content/subject, description of action, state, and mood], [art form, style], [artist/photographer reference if needed], [additional settings such as camera and lens settings, lighting, colors, effects, texture, background, rendering]."}, {"type": "image_url", "image_url": {"url": f"data:image/png;base64,{image_base64}"}}, ], }, ], "temperature": 0.6, } try: response = requests.post(api_url, json=payload, headers=headers) if response.status_code == 200: return response.json() else: st.error(f"API Error: {response.status_code}, {response.text}") return {"error": response.text} except Exception as e: st.error(f"An exception occurred: {e}") return {"error": str(e)} # File uploader for image uploaded_image = st.file_uploader("Upload an image", type=["png", "jpg", "jpeg"]) if uploaded_image and API_KEY: # Convert the image to base64 image = Image.open(uploaded_image) buffered = BytesIO() # Use BytesIO from the io module image.save(buffered, format="PNG") image_base64 = base64.b64encode(buffered.getvalue()).decode() # Show the uploaded image st.image(image, caption="Uploaded Image", use_container_width=True) # Add a button to trigger caption generation if st.button("Generate Prompt", use_container_width=True): st.write("Generating Prompt...") result = generate_caption(image_base64, API_KEY) # Extract and display the generated caption if "error" in result: st.error(f"Error: {result['error']}") else: try: # Extract the caption from the response caption = ( result.get("choices", [{}])[0] .get("message", {}) .get("content", "No caption generated.") ) st.text_area("Prompt Generated", caption, height=200, key="caption_output") except Exception as e: st.error(f"Error processing the response: {e}") else: if not API_KEY: st.warning("Please set the `NEBIUS_API_KEY` environment variable.") else: st.info("Please upload an image.") # Remove the sidebar and center the content st.markdown( """ """, unsafe_allow_html=True )