OpenAI Introduces Cost-efficient GPT-4o Mini Model

OpenAI announced the launch of its new GPT-4o mini model on Thursday. This cost-efficient and energy-saving model aims to make advanced AI technology more affordable and accessible to a wider range of customers. 

Priced competitively at 15 cents per million input tokens and 60 cents per million output tokens, GPT-4o mini is over 60 per cent cheaper than its predecessor, GPT-3.5 Turbo.

Despite its smaller size, GPT-4o mini surpasses its larger counterparts in performance. It excels in chat preferences and achieved an impressive 82 per cent on the Massive Multitask Language Understanding (MMLU) benchmark, which measures a model’s textual intelligence and reasoning across diverse domains. This score places GPT-4o mini ahead of Google’s Gemini Flash, which scored 77.9 per cent and Anthropic’s Claude Haiku, which scored 73.8 per cent.

The introduction of GPT-4o mini is particularly beneficial for companies with limited resources, as smaller language models require less computational power, reducing operational costs while maintaining high performance. Currently, the model supports text and vision in its application programming interface (API), with plans to expand to text, image, video and audio inputs and outputs in the future.

Starting Thursday, GPT-4o mini will be accessible to ChatGPT’s Free, Plus and Team users, replacing GPT-3.5 Turbo. Enterprise users will gain access to the new model next week.