Upload 3 files

Dockerfile

@@ -0,0 +1,13 @@
+FROM python:3.10-slim
+
+WORKDIR /app
+
+# Install dependencies first for layer caching
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy application files
+COPY . .
+
+# Run Chainlit on port 7860 (mandatory for Hugging Face)
+CMD ["chainlit", "run", "main.py", "--port", "7860", "--headless"]

main.py

@@ -0,0 +1,40 @@
+# Import required libraries
+import os  # for environment variables
+import chainlit as cl  # for chatbot interface
+import google.generativeai as genai  # for google genai api
+from dotenv import load_dotenv  # for loading environment variables
+
+# Load environment variables from .env file
+load_dotenv()
+
+# Get Gemini API key from environment variables
+gemini_api_key = os.getenv("GEMINI_API_KEY")
+
+# Configure Gemini API with the API key
+genai.configure(api_key=gemini_api_key)
+
+# Initialize Gemini model
+model = genai.GenerativeModel(model_name="gemini-2.0-flash")
+
+
+# chainlit decorator for when a new chat session starts
+@cl.on_chat_start
+async def handle_chat_start():
+    # Send welcome message to user
+    await cl.Message(content="Hello! how can I help you today?").send()
+
+
+# chainlit decorator for when a new message is received
+@cl.on_message
+async def handle_message(message: cl.Message):
+    # Get the message content from user
+    prompt = message.content
+
+    # Generate response using Gemini model
+    response = model.generate_content(prompt)
+
+    # Extract text from response, or empty string if no text attribute
+    response_text = response.text if hasattr(response, "text") else ""
+
+    # Send response back to user
+    await cl.Message(content=response_text).send()

requirements.txt

@@ -0,0 +1,3 @@
+chainlit
+google-generativeai
+python-dotenv