feat: Enhance multimodal AI agent with improved API key handling and model updates

starter_ai_agents/multimodal_ai_agent/README.md

@@ -1,10 +1,10 @@
 ## 🧬 Multimodal AI Agent
 
-A Streamlit application that combines video analysis and web search capabilities using Google's Gemini 2.0 model. This agent can analyze uploaded videos and answer questions by combining visual understanding with web-search.
+A Streamlit application that combines video analysis and web search capabilities using Google's Gemini 2.5 model. This agent can analyze uploaded videos and answer questions by combining visual understanding with web-search.
 
 ### Features
 
-- Video analysis using Gemini 2.0 Flash
+- Video analysis using Gemini 2.5 Flash/Pro
 - Web research integration via DuckDuckGo
 - Support for multiple video formats (MP4, MOV, AVI)
 - Real-time video processing

starter_ai_agents/multimodal_ai_agent/multimodal_reasoning_agent.py

@@ -1,25 +1,40 @@
 import streamlit as st
 from agno.agent import Agent
+from agno.run.agent import RunOutput
+from agno.media import Image
 from agno.models.google import Gemini
 import tempfile
 import os
 
 def main():
-    # Set up the reasoning agent
-    agent = Agent(
-        model=Gemini(id="gemini-2.0-flash-thinking-exp-1219"), 
-        markdown=True
-    )
-
     # Streamlit app title
     st.title("Multimodal Reasoning AI Agent 🧠")
 
+    # Get Gemini API key from user in sidebar
+    with st.sidebar:
+        st.header("🔑 Configuration")
+        gemini_api_key = st.text_input("Enter your Gemini API Key", type="password")
+        st.caption(
+            "Get your API key from [Google AI Studio]"
+            "(https://aistudio.google.com/apikey) 🔑"
+        )
+
     # Instruction
     st.write(
         "Upload an image and provide a reasoning-based task for the AI Agent. "
         "The AI Agent will analyze the image and respond based on your input."
     )
 
+    if not gemini_api_key:
+        st.warning("Please enter your Gemini API key in the sidebar to continue.")
+        return
+
+    # Set up the reasoning agent
+    agent = Agent(
+        model=Gemini(id="gemini-2.5-pro", api_key=gemini_api_key), 
+        markdown=True
+    )
+
     # File uploader for image
     uploaded_file = st.file_uploader("Upload Image", type=["jpg", "jpeg", "png"])
 
@@ -43,7 +58,7 @@ def main():
                 with st.spinner("AI is thinking... 🤖"):
                     try:
                         # Call the agent with the dynamic task and image path
-                        response = agent.run(task_input, images=[temp_path])
+                        response: RunOutput = agent.run(task_input, images=[Image(filepath=temp_path)])
                         
                         # Display the response from the model
                         st.markdown("### AI Response:")

starter_ai_agents/multimodal_ai_agent/mutimodal_agent.py

@@ -1,5 +1,6 @@
 import streamlit as st
 from agno.agent import Agent
+from agno.run.agent import RunOutput
 from agno.models.google import Gemini
 from agno.media import Video
 import time
@@ -14,15 +15,21 @@ st.set_page_config(
 
 st.title("Multimodal AI Agent 🧬")
 
-# Get Gemini API key from user
-gemini_api_key = st.text_input("Enter your Gemini API Key", type="password")
+# Get Gemini API key from user in sidebar
+with st.sidebar:
+    st.header("🔑 Configuration")
+    gemini_api_key = st.text_input("Enter your Gemini API Key", type="password")
+    st.caption(
+        "Get your API key from [Google AI Studio]"
+        "(https://aistudio.google.com/apikey) 🔑"
+    )
 
 # Initialize single agent with both capabilities
 @st.cache_resource
 def initialize_agent(api_key):
     return Agent(
         name="Multimodal Analyst",
-        model=Gemini(id="gemini-2.0-flash", api_key=api_key),
+        model=Gemini(id="gemini-2.5-flash", api_key=api_key),
         markdown=True,
     )
 
@@ -60,7 +67,7 @@ if gemini_api_key:
                         Provide a comprehensive response focusing on practical, actionable information.
                         """
                         
-                        result = agent.run(prompt, videos=[video])
+                        result: RunOutput = agent.run(prompt, videos=[video])
                         
                     st.subheader("Result")
                     st.markdown(result.content)

starter_ai_agents/multimodal_ai_agent/requirements.txt

@@ -1,3 +1,3 @@
-agno
+agno>=2.2.10
 google-generativeai==0.8.3
 streamlit==1.40.2