OpenAI has officially unveiled GPT-4.1, the successor to last year's GPT-4o model. The company highlights major advancements in coding and instruction-following capabilities with this new model. GPT-4.1 is available in three different versions: the main model, along with the more compact and cost-effective GPT-4.1 mini and GPT-4.1 nano. The new model can process up to 1 million tokens, a striking increase compared to the 128,000 tokens limit of GPT-4o.
Additionally, OpenAI emphasizes that GPT-4.1 is 26% cheaper than its predecessor, making it more accessible. Users can expect significant improvements in the ability to disregard distractions in both long and short contexts. The model also stands out with its multilingual coding capabilities and is now available on platforms like GitHub Copilot.
These developments reflect OpenAI's efforts to maintain its foothold in an increasingly competitive environment. The company is also looking to bolster its market position in light of the anticipated delay of GPT-5's planned launch by several months. With the capabilities offered by this new model, users will be able to work more effectively and efficiently.