promptulate
🚀Lightweight Large language model automation and Autonomous Language Agents development framework. Build your LLM Agent Application in a pythonic way!
What is promptulate?
🚀Lightweight Large language model automation and Autonomous Language Agents development framework. Build your LLM Agent Application in a pythonic way!
Overview of promptulate
What it is
Promptulate is an AI Agent application development framework created by Cogit Lab. It provides a Pythonic development paradigm for building Agent applications. The framework aims to lower the barrier to entry and unify developer consensus by integrating highlights from various open-source frameworks. With Promptulate, users can manipulate components like LLM, Agent, Tool, and RAG with concise code.
Key features
- Pythonic code style: provides a Pythonic SDK calling approach; most tasks can be done with a single
pne.chatfunction. - Model compatibility: supports nearly all types of large models on the market, including OpenAI, Azure, AWS, Google, Mistral, Anthropic, Hugging Face, Ollama, and many more.
- Diverse Agents: offers WebAgent, ToolAgent, CodeAgent, etc., capable of planning, reasoning, and acting.
- Low-cost integration: easily integrates tools from frameworks like LangChain.
- Functions as Tools: any Python function can be directly converted into a tool usable by Agents.
- Lifecycle and Hooks: provides hooks and lifecycle management for Agents, Tools, and LLMs.
- Terminal integration: built-in client support for rapid debugging of prompts.
- Prompt caching: caching mechanism for LLM prompts to reduce repetitive work.
- Powerful OpenAI wrapper:
pne.chatcan replace the OpenAI SDK with enhanced features. - Streamlit component integration: provides out-of-the-box examples and reusable Streamlit components.
Use cases
- Building streamlit chatbots.
- Building gradio chatbots.
- Building math applications with agents.
- Building knowledge maps with Streamlit and pne.
- Building a chatbot to chat with a GitHub repository.
- Creating a multimodal robot agent framework with ROS2 and Promptulate.
- Using pne to compare different models in a playground.
- Using gcop as a Git AI copilot based on Promptulate.
Getting started
Install the framework via pip: shell script pip install -U pne Note: Python version should be 3.8 or higher. After installation, you can use pne.chat() to chat like OpenAI: python import promptulate as pne messages = [ {“role”: “system”, “content”: “You are a helpful assistant.”}, {“role”: “user”, “content”: “Who are you?”}, ] response: str = pne.chat(messages=messages, model=”gpt-4-turbo”) print(response)
License and pricing
The license is Apache-2.0. The project is open source (pricing_type: open_source).
Based on the project's public README and repository metadata.