OpenAI, the artificial intelligence research company behind ChatGPT, has unveiled a new cost-efficient small language model called GPT-4o mini. This model aims to make advanced AI capabilities more accessible to developers and businesses by significantly reducing the costs associated with deploying large language models.
One of the most notable features of GPT-4o mini is its affordability. Priced at just 15 cents per million input tokens and 60 cents per million output tokens, it is more than 60% cheaper than the previous GPT-3.5 Turbo model.
This drastic reduction in costs opens up AI development to a wider range of users, including small businesses and independent developers who may have previously found it prohibitively expensive to integrate AI into their applications.
Read About: OpenAI launches GPT-4o, its newest AI model
Despite its lower cost, GPT-4o mini does not compromise on performance. In fact, it outperforms its predecessors and competitors in several key areas.
The model scores an impressive 82% on the Massive Multitask Language Understanding (MMLU) benchmark, surpassing Google’s Gemini Flash (77.9%) and Anthropic’s Claude Haiku (73.8%) in reasoning tasks.
GPT-4o mini also boasts a 128,000 token context window, allowing it to handle large volumes of text effectively. It is reported to be twice as fast as GPT-4 Turbo, with a rate limit five times higher than that model.
In terms of capabilities, the model currently supports text and vision inputs and outputs, with plans to integrate audio and video functionalities in the future. This multimodal approach enables a wide range of applications, from image captioning and visual question answering to generating images from textual descriptions.
GPT-4o mini is accessible through OpenAI’s API, including the Chat Completions API, Assistants API, and Batch API. It is also available to users of the ChatGPT web and mobile applications, making it readily available to a broad audience.
Related: ChatGPT: 10 things you should not use it for
The model is particularly well-suited for high-volume tasks, such as customer support and content generation, where its speed and efficiency are critical. Its cost-effectiveness also makes it an attractive option for budget-conscious projects and applications.
OpenAI emphasizes its commitment to responsible AI development with GPT-4o mini. Safety measures are ingrained throughout the model’s creation process, including filtering training data to prevent the incorporation of harmful content and utilizing reinforcement learning with human feedback to ensure accurate and reliable responses.
Furthermore, GPT-4o mini incorporates an “instruction hierarchy” designed to resist attempts to manipulate the model’s behaviour.
