GPT-4.1 is the flagship model delivering significant improvements in coding, instruction following, and long context processing. It creates more functional and aesthetically pleasing web applications, with human evaluators preferring its websites 80% of the time over GPT-4o.

Through efficiency improvements, GPT-4.1 is 26% less expensive than GPT-4o for median queries, with a 75% prompt caching discount for repeated contexts. Long context requests come at no additional cost beyond standard per-token pricing.

‍

This latest iteration in the GPT series comes with significant enhancements that improve performance in areas such as coding, instruction following, and managing long context.