OpenAI Releases GPT-4.1 with Enhanced Capabilities
On April 14, 2025, OpenAI unveiled GPT-4.1, the latest iteration in its series of AI language models, bringing significant enhancements in coding abilities, instruction adherence, and long-context comprehension. This release includes three variants: GPT-4.1, GPT-4.1 Mini, and GPT-4.1 Nano, each tailored to different performance and cost requirements. Notably, GPT-4.1 supports a context window of up to one million tokens, enabling it to process extensive inputs effectively.
The GPT-4.1 models demonstrate substantial improvements in coding tasks, outperforming their predecessors on benchmarks such as SWE-bench Verified, where GPT-4.1 achieved a 54.6% score—marking a 21.4% absolute improvement over GPT-4o. These advancements position GPT-4.1 as a leading model for coding applications. Additionally, the models have been optimized for real-world use based on direct developer feedback, enhancing their performance in frontend coding, adherence to response structures, and consistent tool usage.
In terms of accessibility and cost-efficiency, GPT-4.1 is priced at $2 per million input tokens and $8 per million output tokens. The Mini and Nano variants offer more budget-friendly options, catering to developers with varying needs. OpenAI plans to discontinue the GPT-4.5 preview in the API by July, reflecting confidence in GPT-4.1's superior performance and cost-effectiveness. CEO Sam Altman emphasized the focus on real-world utility, noting strong satisfaction from developers utilizing the new models.
