Streamlit UI Interface and Other Small Changes

.env.example

@@ -0,0 +1,3 @@
+DEEPSEEK_API_KEY=your_deepseek_api_key_here 
+OPENROUTER_API_KEY=your_openrouter_api_key_here 
+ANTHROPIC_API_KEY=your_anthropic_api_key_here 

README.md

@@ -24,6 +24,25 @@ This approach ensures more thoughtful, contextually aware, and reliable response
 - 🧠 **Reasoning Visibility**: Toggle visibility of the AI's thinking
 - 🔄 **Context Awareness**: Maintains conversation context for more coherent interactions
 
+## 🖥️ Streamlit Interface
+
+The new Streamlit interface offers a modern, user-friendly way to interact with RAT:
+
+- 💻 **Web-Based UI**: Clean, intuitive interface with dark theme support
+- 🔄 **Live Streaming**: Watch reasoning and responses being generated in real-time
+- 🤖 **Advanced Model Management**:
+  - Manually enter models with automatic provider prefix detection
+  - Fetch top weekly models directly from OpenRouter (may take 1-2 minutes)
+  - Select models from a dropdown menu of popular options
+- 📊 **Token Tracking**: Transparent display of token usage for each response:
+  - Input tokens (question)
+  - Reasoning tokens (thinking process)
+  - Response tokens (final answer)
+  - Total tokens consumed
+- 🧠 **Reasoning Toggles**: Show or hide the detailed reasoning process for each response
+- 📜 **Conversation History**: Full chat history with metadata for each interaction
+- ⚙️ **Customizable Settings**: Adjust interface preferences via sidebar controls
+
 ## ⚙️ Requirements
 
 • Python 3.11 or higher  
@@ -73,6 +92,17 @@ This will install RAT as a command-line tool, allowing you to run it from anywhe
    - Type "reasoning" to show/hide the thinking process
    - Type "quit" to exit
 
+4. Launch the Streamlit interface:
+   ```bash
+   streamlit run rat/ui.py
+   ```
+
+5. Using the Streamlit interface:
+   - Ask questions in the chat input at the bottom
+   - View detailed token usage for each response
+   - Toggle reasoning visibility in the sidebar
+   - Change models manually or select from top weekly models
+   - Clear chat history as needed
 
 
 ## 🚀 Versions
@@ -92,6 +122,13 @@ Run it using:
 uv run rat-claude.py
 ```
 
+### Streamlit Interface (ui.py)
+A modern web-based interface with real-time streaming, token tracking, and model selection.
+Run it using:
+```bash
+streamlit run rat/ui.py
+```
+
 
 ## 🤝 Contributing
 
@@ -116,5 +153,4 @@ GitHub: https://github.com/yourusername/rat
 
 ## Star History
 
-[![Star History Chart](https://api.star-history.com/svg?repos=Doriandarko/RAT-retrieval-augmented-thinking&type=Date)](https://star-history.com/#Doriandarko/RAT-retrieval-augmented-thinking&Date)
-
+[![Star History Chart](https://api.star-history.com/svg?repos=Doriandarko/RAT-retrieval-augmented-thinking&type=Date)](https://star-history.com/#Doriandarko/RAT-retrieval-augmented-thinking&Date)

pyproject.toml

@@ -12,6 +12,9 @@ dependencies = [
     "pyperclip>=1.9.0",
     "python-dotenv>=1.0.1",
     "rich>=13.9.4",
+    "selenium>=4.30.0",
+    "streamlit>=1.44.0",
+    "tiktoken>=0.9.0",
 ]
 
 [project.scripts]

rat-claude.py

@@ -175,7 +175,7 @@ def main():
                 continue
 
             reasoning = chain.get_deepseek_reasoning(user_input)
-            claude_response = chain.get_claude_response(user_input, reasoning)
+            claude_response = chain.get_claude_response(user_input, reasoning)  # noqa: F841
 
         except KeyboardInterrupt:
             continue