OpenAI has officially launched GPT-4.1 and GPT-4.1 mini across its ChatGPT platform, introducing faster performance, sharper reasoning, and improved coding capabilities. The rollout is now live for Plus, Pro, and Team subscribers, while free-tier users can access the lighter GPT-4.1 mini model.
The announcement, made on X (formerly Twitter), comes shortly after the model’s April debut through OpenAI’s developer API. With the release, GPT-4.0 mini has been retired, making way for a more streamlined and responsive experience across user tiers.
According to OpenAI spokesperson Shaokyi Amdo, GPT-4.1 offers clear improvements in code generation and debugging, making it especially attractive to software developers. It also handles complex reasoning tasks more effectively and demonstrates tighter instruction-following than GPT-4o, OpenAI’s previous flagship model.
Also Read: OpenAI Unveils GitHub Connector to Boost Code Research in ChatGPT
However, the launch sparked some criticism due to the absence of a full safety report, a detail traditionally released with major updates. OpenAI clarified that GPT-4.1 isn’t categorized as a “frontier model,” which exempts it from certain transparency requirements. Johannes Heidecke, OpenAI’s Head of Safety Systems, stated that the model doesn’t introduce new modalities or outpace the intelligence level of existing models like o3.
In an effort to improve accountability, OpenAI has also launched a new Safety Evaluations Hub to share more insights into its internal safety testing processes. The update lands as competition in the AI field intensifies—Google, for instance, has added GitHub integration to its Gemini chatbot. Meanwhile, OpenAI is reportedly in talks to acquire Windsurf, an AI coding platform, for $3 billion.
