🤖 PromptForge

A minimal, fast, multi-model AI assistant you control

Switch models • Stream responses • Keep it simple

---

✨ Features

• 💬 Chat-style interface (like ChatGPT, Gemini)
• ⚡ Streaming responses
• 🔄 Switch between models:
  • google/gemma-3-27b-it
  • nvidia/nemotron-3-super-120b-a12b:free
  • You can add more...
• 🧹 Clear chat option

---

⚙️ Setup Instructions

1. Clone the repository

git clone <your-repo-url>
cd <your-repo-folder>

---

2. Create virtual environment

python -m venv .venv

Activate it:

Windows

.venv\Scripts\activate  

Mac/Linux

source .venv/bin/activate  

---

3. Install dependencies

pip install streamlit pyautogen python-dotenv autogen-ext[openai]

---

4. Create .env file

OPENROUTER_API_KEY=your_api_key_here  
OPENROUTER_BASE_URL="https://openrouter.ai/api/v1"  

---

5. Run the app

streamlit run app.py

---

🧠 Models Used

This app uses models via OpenRouter:

• Gemma 3 (27B) → Fast, lightweight
• Nemotron 3 Super (120B) → Strong reasoning

---

📌 Notes

• Some models may not stream token-by-token (depends on provider)
• Responses may include artifacts like TERMINATE (model-specific)

---

🚀 Future Improvements

• Code review mode
• File upload support
• Multi-agent workflows
• Download responses / code