APrompt 🤖

An automated prompt engineering tool for Large Language Models (LLMs), designed for universal domain adaptation.

✨ Features

• 💰 Ultra-low iteration cost ($0.1 level)
• 🏷️ Label-free approach - no training data needed
• 📊 Few-shot learning with minimal examples
• ⚡ Super simple template configuration
• 🌐 Multi-language support

🚀 Quick Start

1. Configure Your API Key ⚙️

Create a configuration file config.yaml:

openai:
  api_key: " "
  base_url: " "

2. Define Your Prompt Template 📝

Create a prompt template file settings/task_name.yaml:

prompt: |
  solve question.

requirements: |
  ...

count: None

faq:
  - question: |
      ...
    answer: |
      ...

  - question: |
      ...
    answer: |
      ...

3. Implement the Optimizer 🔧

Use main.py to execute:

from script.optimizer import Optimizer

if __name__ == "__main__":

    optimizer = Optimizer(
        optimized_path="workspace",
        initial_round=1,
        max_rounds=10,
        template="task_name.yaml",
        name="Task",
        optimize_model={"name": "claude-3-5-sonnet-20240620", "temperature": 0.7},
        execute_model={"name": "gpt-4o-mini", "temperature": 0},
        evaluate_model={"name": "gpt-4o-mini", "temperature": 0.3},
        iteration=True,
    )

    optimizer.optimize()

📄 License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

---

Made with ❤️ by the MetaGPT and AFlow Team