OpenAI has launched GPT-4o mini, a new, streamlined version of its AI model designed to be more cost-effective and accessible. This new model is set to replace GPT-3.5 Turbo in ChatGPT, offering users enhanced capabilities at a lower cost.

GPT-4o mini is touted as the most functional and economically efficient small model available today. It supports both text and computer vision via API, with plans to handle text, image, video, and audio inputs and outputs in the near future. Initially available to developers, it will soon be accessible to general users as well.

The model boasts an impressive context window of 128,000 tokens and supports up to 16,000 output tokens per request. Its knowledge base is current up to October 2023. In terms of performance, GPT-4o mini surpasses other small models in several benchmarks.

OpenAI Releases a Faster GPT-4o Model That Is Free for All Users
The GPT-4o model is designed to handle real-time interactions across multiple formats—whether it’s responding to text inquiries, engaging in voice conversations, or recognizing and analyzing images through a camera.
Recorded News — Tech NewsBradley Rhodes

One of the standout features of GPT-4o mini is its energy efficiency. It costs 15 cents per million input tokens and 60 cents per million output tokens, making it 60% cheaper than GPT-3.5 Turbo. This significant reduction in cost aims to make AI technology more accessible to a wider range of developers and users.

Safety remains a priority for OpenAI. GPT-4o mini incorporates a "hierarchy of instructions" system to prioritize safety instructions and prevent misuse. This model also features advanced safety measures to resist jailbreaks and prompt injections, ensuring more reliable and secure usage.

GPT-4o mini is now available for developers via the API, and it will be integrated into the free version of ChatGPT soon.