OpenAI Introduces GPT-4.1 With Major Performance Upgrades

OpenAI has released a new family of AI models, led by GPT-4.1, along with smaller versions called GPT-4.1 mini and GPT-4.1 nano. These models offer major improvements in coding, instruction following, and long-context understanding.

Available only through OpenAI’s API, GPT-4.1 outperforms the previous GPT-4o model across all key areas. It also surpasses GPT-4.5, especially in programming tasks. According to OpenAI, GPT-4.1 shows a 21% improvement over GPT-4o and a 27% gain over GPT-4.5 in coding benchmarks.

In addition, GPT-4.1 supports up to 1 million tokens. This means the model can process and understand much longer texts, which benefits developers building complex AI tools. The models also come with updated knowledge up to June 2024, offering more relevant and current responses.

OpenAI noted that the new models run at a much lower cost compared to GPT-4.5. As a result, the company plans to remove the GPT-4.5 preview from its API in July. GPT-4.1 provides equal or better performance, making the older version unnecessary.

CEO Sam Altman said on social media that while benchmarks were important, the focus was on real-world use. He added that early developer feedback has been very positive.

With the launch of GPT-4.1, OpenAI aims to deliver faster, smarter, and more cost-effective AI solutions for a wide range of applications.