OpenAI Rolls Out GPT-4.1 and GPT-4.1 Mini Models in ChatGPT, Replacing GPT-4o Mini
San Francisco | May 17, 2025 — In a significant update to its AI offerings, OpenAI has announced that GPT-4.1 and GPT-4.1 Mini are now available for all users within ChatGPT, marking a major enhancement in performance and versatility for the widely-used AI chatbot. Previously, these models were only accessible via the OpenAI API after their launch a month ago.
The latest update signals OpenAI’s intention to streamline its ChatGPT model lineup while optimizing for performance, safety, and task-specific strengths. As part of this transition, OpenAI has begun phasing out the GPT-4o Mini model from ChatGPT, citing improvements in both speed and capability with the new GPT-4.1 models.
GPT-4.1: Optimized for Performance and Coding
The GPT-4.1 model has been especially praised for its capabilities in coding, web development, and structured problem-solving, outperforming earlier versions in both response time and precision. It’s designed to cater to developers, students, and tech professionals looking for a highly responsive and accurate coding assistant.
“GPT-4.1 builds on the safety work and mitigations developed for GPT-4o. Across our standard safety evaluations, GPT-4.1 performs at parity with GPT-4o, showing that improvements can be delivered without introducing new safety risks,” said Johannes Heidecke, OpenAI’s Head of Safety Systems, in a statement on X (formerly Twitter).
Heidecke noted that although GPT-4.1 represents a step forward in performance and robustness, it does not surpass GPT-4o in general intelligence or interactive capabilities, as GPT-4.1 does not introduce any new modalities or interface mechanisms. GPT-4o, which had been referred to as “o3,” was previously recognized for its enhanced multi-modal interaction—combining text, vision, and speech.
GPT-4.1 Mini: Lightweight Model for Everyday Use
Alongside the main GPT-4.1 model, OpenAI has introduced GPT-4.1 Mini, a lightweight version designed to deliver faster responses and lower compute usage while maintaining a strong level of capability for casual and professional users alike. It serves as a powerful alternative for those who may not need the full extent of GPT-4.1’s advanced functionalities but still seek a reliable AI assistant.
OpenAI’s move comes just two weeks after the company quietly retired the GPT-4 model from ChatGPT on April 30, signaling a broader effort to consolidate and refine its AI offerings based on evolving user needs and technical progress.
Evolution of ChatGPT Models: A Strategic Shift
The evolution from GPT-4 and GPT-4o Mini to GPT-4.1 and GPT-4.1 Mini represents a strategic recalibration of OpenAI’s product roadmap. By unifying performance with safety and task specialization, the company continues to prioritize user experience and scalable deployment.
Since its inception, ChatGPT has undergone several iterations, with each model aimed at striking a balance between intelligence, ethical safety, and real-world utility. GPT-4.1 appears to maintain this trend, boasting faster computation times and expanded support for programming and analytical tasks—areas where generative AI is increasingly in demand.
OpenAI has yet to confirm if additional models or capabilities—such as multi-modal interaction, voice support, or image understanding—will be integrated into GPT-4.1 in the near future. However, with the company actively experimenting across multiple domains, users can likely expect further enhancements throughout 2025.
The introduction of GPT-4.1 into ChatGPT is now live for users on supported plans, and OpenAI is expected to release more detailed documentation and updates on its developer and community forums soon.