Openai will soon retire from the GPT-4, an AI model released more than two years ago, according to Changelog, posted Thursday.
On April 30, the GPT-4 will be “fully replaced” by the GPT-4O, the current default model of CHATGPT, Openai said. GPT-4 will remain available through OpenAI’s API.
“According to the head evaluation, [GPT-4o] Openai wrote in Changelog. “The recent upgrades have further improved the teaching, problem solving and conversation flow of GPT ‑ 4o and become a natural successor to GPT ‑ 4.”
GPT-4 was deployed in March 2023 for ChatGPT on the web and Microsoft’s Copilot Chatbot. Several versions of GPT-4 have multimodal features, allowing you to understand both images and text. This is the first of the widely deployed OpenAI model.
Openai CEO Sam Altman said the GPT-4 is reportedly large in size and would cost more than $100 million to train. It was taken over by the GPT-4 Turbo in November 2023, making it a faster and cheaper model.
The GPT-4 is one of the models at the heart of the copyright dispute between Openai and publishers, including The New York Times. Publishers claim that Openai trained GPT-4 on data without knowledge or consent. Openai claims that the doctrine of fair use protects it from liability.
Future retirements of GPT-4 could follow the release of new models at ChatGPT. According to reverse engineer Tibor Blaho, Openai is preparing a family of models called GPT-4.1 – GPT-4.1-MINI, GPT-4.1-NANO and GPT-4.1.
Source link
