Update utils/gemini_utils.py

utils/gemini_utils.py

@@ -1,5 +1,7 @@
 import google.generativeai as genai
 import streamlit as st
+from PIL import Image
+import io
 
 # Set up the API key
 GOOGLE_API_KEY = st.secrets["GOOGLE_API_KEY"]
@@ -14,37 +16,25 @@ generation_config = {
 }
 
 model = genai.GenerativeModel(
-    model_name="gemini-1.5-pro",
+    model_name="gemini-pro-vision",
     generation_config=generation_config,
 )
 
-def upload_to_gemini(image_bytes, mime_type="image/png"):
-    """Uploads the given image bytes to Gemini."""
-    try:
-        file = genai.types.GenerativeContentBlob(data=image_bytes)
-        return file
-    except Exception as e:
-        st.error(f"Error uploading file: {e}")
-        return None
-
 def get_meme_template_description(image_bytes):
     """Get a description of the meme template from the image bytes."""
     try:
-        file = upload_to_gemini(image_bytes)
-        if file:
-            prompt = """
-            You are an AI specialized in describing meme templates. Please provide a detailed description of the meme template in this image. Include:
-            1. The name of the meme template (if it's a well-known one)
-            2. A description of the visual layout (number of panels, arrangement)
-            3. Key visual elements in each panel
-            4. The typical use or purpose of this meme template
-            5. Where text is usually placed in this template
+        image = Image.open(io.BytesIO(image_bytes))
+        prompt = """
+        You are an AI specialized in describing meme templates. Please provide a detailed description of the meme template in this image. Include:
+        1. The name of the meme template (if it's a well-known one)
+        2. A description of the visual layout (number of panels, arrangement)
+        3. Key visual elements in each panel
+        4. The typical use or purpose of this meme template
+        5. Where text is usually placed in this template
 
-            Format your response as a concise paragraph.
-            """
-            response = model.generate_content([file, prompt])
-            return response.text
-        else:
-            return "Failed to upload image."
+        Format your response as a concise paragraph.
+        """
+        response = model.generate_content([prompt, image])
+        return response.text
     except Exception as e:
         return f"Error processing image: {e}"