OpenAI, the Microsoft-backed leader in AI software, is set to introduce GPT-4o mini, a new cost-efficient small AI model. This move aims to make its technology more affordable and less energy-intensive for its AI customers.
The GPT-4o mini will retail at 15 cents per million input tokens and 60 cents per million output tokens, making it over 60 percent cheaper than GPT-3.5 Turbo, according to OpenAI. Despite its smaller size, the GPT-4o mini outperforms the GPT-4 model in chat preferences and scored 82 percent on the Massive Multitask Language Understanding (MMLU) benchmark. This score is notably higher than Google’s Gemini Flash (77.9 percent) and Anthropic’s Claude Haiku (73.8 percent), showcasing its superior textual intelligence and reasoning capabilities across various domains.
Smaller language models like the GPT-4o mini require less computational power, offering an affordable option for companies with limited resources to deploy generative AI in their operations. The new model currently supports text and vision in its API, with plans to expand support to text, image, video, and audio inputs and outputs in the future.
Starting Thursday, ChatGPT’s Free, Plus, and Team users will have access to GPT-4o mini, replacing GPT-3.5 Turbo. Enterprise users will gain access starting next week. The model contains knowledge up to October 2023, providing up-to-date information and capabilities.
As OpenAI continues to innovate, major players like Meta Platforms, Amazon, and Google are also striving to increase their market share in the competitive AI space.
