OpenAI has released GPT-3, a state-of-the-art language model made up of 175 billion parameters. In this video, I'll create a simple tutorial on how you can use OpenAI's API to use the GPT-3 model.

The previous OpenAI GPT model that is GPT-2 had 1.5 billion parameters and was the biggest model back then. GPT-3 can write poetry, translate text, chat convincingly, and answer abstract questions.

Microsoft announced the first features of a customer product driven by GPT-3, a strong natural language model built by OpenAI that would let users to build apps without having to know how to write computer code or formulae, at its Build developers conference.

GPT-3 will be included in Microsoft Power Apps, a low-code app creation platform that enables everyone from "citizen developers" with no coding experience to professional developers with extensive programming knowledge to create apps to improve corporate efficiency or procedures. Apps to analyze non-profit gift donations, coordinate travel for COVID-19, and limit the amount of overtime necessary to service wind turbines are all examples.

For example, the new AI-powered capabilities would allow a developer working on an e-commerce platform to explain a programming objective in conversational languages, such as "identify items whose names begin with the letter 'kids.'" A fine-tuned GPT-3 model then provides options for converting the command into a Microsoft Power Fx formula, the Power Platform's open-source programming language, such as "Filter('BC Orders' Left('Product Name',4)="Kids").

It's one of the first examples of how GPT-3, which runs on Microsoft Azure and is driven by Azure Machine Learning, can address real-world business problems on a large scale, according to Microsoft. It's also one of the first internal deployments of Microsoft's new managed endpoints feature.