update REAME

README.md

@@ -9,59 +9,60 @@ app_file: app.py
 pinned: false
 models:
   - google/gemma-2b
-  - google/gemma-2b-it
+  - google/gemma-3b
   - google/gemma-7b
-  - google/gemma-7b-it
 tags:
   - gemma
   - chat
   - language-model
   - code-generation
-short_description: A chat interface for Google's Gemma models.
+short_description: A friendly chat interface for Google's Gemma models.
 ---
 
 # Gemma Chat Demo
 
-An interactive chat application powered by Google's Gemma models using Hugging Face and Streamlit.
-
-HuggingFace Spaces Link: https://huggingface.co/spaces/aadya1762/GemmaDemoSt2
+A simple and friendly chat application powered by Google's Gemma AI models.
 
 ## Features
 
-- 🔐 Hugging Face authentication for accessing Gemma models
-- 🤖 Support for multiple Gemma model variants (2B, 7B, base and instruction-tuned)
-- 🔄 Task selection for different conversation types:
-  - Question Answering
-  - Text Generation
-  - Code Completion
-- 💬 Clean chat interface with message history
-- 🧹 Option to clear chat history
+- 💬 Chat with Google's Gemma AI models right on your computer
+- 🔄 Choose between different models:
+  - gemma-3b: Balanced performance (recommended)
+  - gemma-2b: Faster responses
+  - gemma-7b: More capable but requires more resources
+- 🧩 Different chat modes:
+  - Question Answering: Get helpful answers to your questions
+  - Text Generation: Continue any text or story
+  - Code Completion: Get help with coding
+- 🖥️ Simple, easy-to-use interface
+- 📚 No complex setup needed
 
-## Usage
+## Getting Started
 
-1. Log in with your Hugging Face token in the sidebar
-2. Select your preferred Gemma model
-3. Choose a task type for your conversation
-4. Start chatting with the model!
+1. Install the required packages:
+   ```
+   pip install -r requirements.txt
+   ```
 
-## Requirements
+2. Run the application:
+   ```
+   python app.py
+   ```
 
-- Python 3.12
-- Hugging Face account with access to Gemma models
-- Dependencies listed in requirements.txt
+3. When the application starts:
+   - Select a model from the dropdown
+   - Choose a chat mode
+   - Start typing and chatting!
 
-## Running Locally
+## Requirements
 
-```bash
-# Install dependencies
-pip install -r requirements.txt
+- Python 3.11 or newer 
+- Internet connection for first-time model download
 
-# Run the application
-python app.py
-```
+## Tips
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+- For code completion, start by typing the beginning of your code
+- The first time you select a model, it may take a moment to download
+- You can switch between models anytime during your conversation
 
-## Limitations
-- CPU Implementation is very slow (a simple code completion can take around 10 minutes).
-- Limit Scope of Optimization (`torch.compile` takes several minutes to compile & recompiles frequently)
+Enjoy chatting with Gemma!