OpenAI unveils GPT 4.1 with enhanced coding capabilities

OpenAI has announced the release of three new models in its GPT-4.1 series—GPT-4.1, GPT-4.1 mini, and GPT-4.1 nano—through the OpenAI API. These models are designed to deliver significant improvements in coding tasks, instruction following, and long-context comprehension. Notably, they are not yet available on ChatGPT.​

The flagship GPT-4.1 model supports up to one million tokens of context, enhancing its ability to understand and generate long-form content. Benchmark tests reveal that GPT-4.1 outperforms its predecessors, showing a 21.4% improvement over GPT-4o and a 26.6% improvement over GPT-4.5 in coding tasks on the SWE-bench Verified benchmark.

For developers seeking cost-effective solutions, GPT-4.1 mini offers comparable performance to GPT-4o with reduced latency and an 83% lower cost. GPT-4.1 nano, the most streamlined model in the series, is optimized for tasks like classification and autocompletion, providing the fastest and most affordable option.​

OpenAI has indicated that many of the enhancements found in GPT-4.1 have already been integrated into the ChatGPT version of GPT-4o, with additional features to be added in the future. All GPT-4.1 models have a knowledge cutoff date of June 2024.

With the introduction of the GPT-4.1 series, OpenAI is phasing out GPT-4.5 in the API, as the new models offer similar functionalities at a lower cost. Pricing for GPT-4.1 is set at $2 per million input tokens and $8 per million output tokens.GPT-4.1 mini is priced at $0.40 per million input tokens and $1.60 per million output tokens, while GPT-4.1 nano is available at $0.10 per million input tokens and $0.40 per million output tokens.

This strategic release underscores OpenAI's commitment to providing scalable and efficient AI solutions tailored to diverse coding and development needs.​