OpenAI to Retire GPT-4 from ChatGPT, Replacing It with Advanced GPT-4o Model
OpenAI has announced plans to officially retire its GPT-4 model from ChatGPT at the end of April, according to an update published in the company’s changelog, as reported by TechCrunch. Beginning April 30, GPT-4 will be entirely replaced by the newer GPT-4o, which is currently the standard model in ChatGPT. However, users will still be able to access GPT-4 through OpenAI’s API.
In the update, OpenAI said, “In head‑to‑head evaluations, [GPT-4o] consistently surpasses GPT‑4 in writing, coding, STEM, and more.”
He added, “Recent upgrades have further improved GPT‑4o’s instruction following, problem solving, and conversational flow, making it a natural successor to GPT‑4.”
GPT-4’s Legacy and Evolution
First launched in March 2023, GPT-4 made waves as the core model behind both ChatGPT and Microsoft’s Copilot chatbot. It introduced multimodal capabilities, enabling it to process both text and images which is a first for OpenAI’s widely deployed models.
OpenAI CEO Sam Altman previously revealed that GPT-4 was a significant investment, reportedly costing over $100 million to train. In November 2023, OpenAI followed up with GPT-4 Turbo, an iteration designed to deliver faster processing at lower operational costs.
Despite its technological advancements, GPT-4 has been entangled in legal controversies. Several publishers, including The New York Times, have filed lawsuits claiming that OpenAI used their copyrighted materials without permission to train the model. OpenAI maintains that its practices fall under fair use provisions.
Also read: Verizon’s AI Assistant, Powered by Google Cloud, Boosts Sales and Cuts Call Times
What’s Next for ChatGPT Users
According to TechCrunch, the phase-out of GPT-4 from ChatGPT aligns with the expected introduction of new models. Reverse engineer Tibor Blaho recently uncovered that OpenAI is preparing to launch a series of GPT-4.1 models, including GPT-4.1-mini, GPT-4.1-nano, and a standard GPT-4.1. Additionally, OpenAI appears to be working on a new line of reasoning models, with the "o3" model already announced last December, and another named o4-mini in development.
