OpenAI has unveiled GPT-4o mini, a compact artificial intelligence model designed for developers, promising cost-efficiency that surpasses OpenAI’s best value. This is expected to make AI applications more accessible by offering high performance at a lower cost.
The GPT-4o mini is priced at 15 cents per 1 million input tokens and 60 cents per 1 million output tokens, making it 60% cheaper than the GPT-3.5 Turbo. Currently supporting text and vision APIs, it is poised to incorporate data ranging from text, image, video, and audio in the future, with a maximum output of 16k tokens per request. Data updates are scheduled until October 2023.
OpenAI claims that performance tests reveal GPT-4o mini outperforming smaller AI models like Gemini Flash or Claude Haiku in MMLU, MGSM, MMMU tests, as well as surpassing the capabilities of the GPT-3.5 Turbo.
GPT-4o mini will be available to all subscription packages starting today, replacing the GPT-3.5 Turbo. Enterprise customers can expect access to the model starting next week.
TLDR: OpenAI introduces GPT-4o mini, a cost-effective AI model with enhanced performance, offering developers access to a wide range of AI applications.
Leave a Comment