OpenAI has announced that it is phasing out GPT-4.5 from its developer API in favor of the new GPT-4.1 model. At launch, OpenAI described GPT-4.5 as its best and most capable model, in part because it was a more natural conversationalist and could mimic some of the concepts of emotional intelligence. Despite its name, GPT-4.1 is supposed to be better and more efficient.
OpenAI states that GPT-4.1 is available exclusively to developers using the OpenAI API and comes in three sizes: GPT-4.1, GPT-4.1 mini, and GPT-4.1 nano. This means that if you don’t find it as an option in the public ChatGPT interface, but at some point you will be able to interact with an agent that uses the model enhancements. GPT-4.1 should be better at encoding and “understanding long context,” according to OpenAI, with support for “up to one million context tokens” and world knowledge by June 2024.
Important to the decision to phase out the previous version of GPT-4.5 is that GPT-4.1 is also cheaper to use. OpenAI claims that this model is “26% cheaper than GPT-4o for median requests.” The company expects developers to use GPT-4.1 to create agents that can run in software engineering environments, which is one of the most popular uses of generative AI. Although GPT-4.5 will eventually be removed from the OpenAI API, a preview of the model will still be available in ChatGPT if you don’t want to play with it.
OpenAI says that the GPT-4.1 family is available for use today through the company’s API. GPT-4.5 will be fully phased out in three months, on July 14, 2025.
