Reply to thread

Message

[QUOTE="Munyaradzi Mafaro, post: 33672, member: 636"]
OpenAI released GPT-4.1, replacing last year's GPT-4 model. The company claims this update performs better "in just about every dimension," especially for coding tasks. It also handles instructions more effectively.

Developers can access GPT-4.1 alongside smaller versions - Mini for affordability and Nano, described as their "smallest, fastest, and cheapest" option. All three models process up to one million tokens of context, far exceeding GPT-4o's 128,000-token limit. The new model costs 26 percent less than its predecessor, an important factor after DeepSeek introduced its efficient competitor.

OpenAI plans to remove the two-year-old GPT-4 from ChatGPT on April 30, noting that recent GPT-4 improvements make it a natural replacement. The company will also end the GPT-4.5 API preview service on July 14, as GPT-4.1 delivers similar or better results more efficiently.

CEO Sam Altman announced GPT-5's postponement on X, explaining that integration proved "harder than we thought." The delay pushes the launch several months past the expected May timeline. Recent ChatGPT web code references the upcoming o3 reasoning model and o4 mini reasoning model, suggesting another release soon.
[/QUOTE]

Name