The open-source Project Jupyter, used by millions for data science and machine learning, has released Jupyter AI, a free tool bringing powerful generative AI capabilities to Jupyter notebooks.
The tool includes %%ai magic commands for notebooks. The tool also includes a chat interface called Jupyternaut to leverage large language models like GPT-3, Claude, and Amazon Titan to generate, explain, and code directly within JupyterLab and Jupyter notebooks. Additionally, users can use the "/generate" command to produce entire notebooks from text prompts.
Amazon also announced an Amazon CodeWhisperer Jupyter extension. Apart from these generative AI extensions, AWS has introduced new Jupyter extensions to build, train, and deploy ML at scale. These include notebooks scheduling, SageMaker open-source distribution, and the Amazon CodeGuru Jupyter extension.
The innovative technology of Jupyter AI not only helps simplify tasks and save the developer's time but also increases their ability to tackle new difficulties. Gleb Tsipursky wrote in Forbes, "Generative AI's coding ability will automate repetitive tasks, increase productivity, and enhance collaboration between humans and AI, creating a symbiotic relationship that will redefine the programming landscape".
The use of generative AI in software development has been steadily increasing. Research by McKinsey Digital shows how generative AI like Jupyter AI accelerates updates to existing codes and provides a head start to the first draft of a new code.
Jupyter senior front-end engineer Jason Weill has pointed out the intention of creating ethical, socially responsible AI with data protection in mind, and that is evident in features such as the freedom of users to select their favourite LLM, embedding model, and vector database to meet their individual needs. Jupyter AI aims to protect the privacy of user data and only communicates with LLMs upon request; data is never accessed or transmitted without the user's consent.
Users can use pip to install JupyterLab (version 3 or 4) to begin using Jupyter AI. The extension can be found here on GitHub.