💬 ChatOrbit

ChatOrbit is a versatile and interactive AI chatbot built with Streamlit. It enables users to chat with various powerful large language models (LLMs) — including Gemini 2.0, DeepSeek R1, Meta-LLaMA 3.1, and Mixtral-8x7B — all from one clean interface. Choose your preferred model, ask anything, and get smart, fast answers.

---

🚀 Features

• 🧠 Multi-Model Support: Switch between Gemini and Together AI models seamlessly.
• 💬 Chat History: Preserves chat messages per session.
• 🔁 Reset Chat: Clear the conversation with a single click.
• 🎨 Streamlit UI: Clean, reactive UI built with Streamlit for simplicity and speed.
• ⚙️ Async Execution: Uses asyncio for efficient handling of AI requests.

---

🛠️ Tech Stack

• Python 3.10+
• Streamlit
• AsyncOpenAI wrapper
• Together AI & Gemini API
• Dotenv for API key management

---

🔑 Requirements

Make sure you have the following API keys:

• GEMINI_API_KEY – for accessing Gemini models
• AI_TOGETHER_API_KEY – for accessing Together AI models

Create a .env file in the root directory:

`env GEMINI_API_KEY=your_gemini_api_key AI_TOGETHER_API_KEY=your_together_ai_key

🧪 Models Included Model Name Provider Description Gemini 2.0 Flash Google Gemini Fast, general-purpose model DeepSeek R1 Together AI LLaMA-based distillation Meta-LLaMA 3.1 8B Together AI Instruction-tuned Meta model Mixtral-8x7B Instruct Together AI Sparse mixture-of-experts model

🏷️ Topics streamlit chatbot llm openai together-ai gemini-api python conversational-ai asyncio

📄 License MIT License © 2025 [MuhammadUsmanGM]

🙌 Contributing Pull requests are welcome. For major changes, please open an issue first to discuss what you’d like to change.

📸 Model Output Screenshots

🤖 Gemini 2.0 Output

🔍 DeepSeek R1 Output

🦙 Meta-LLaMA 3.1 Output

🔀 Mixtral-8x7B Output