OpenAI unveils GPT-4o mini, its most affordable and efficient AI model yet
OpenAI has released GPT-4o mini, a new AI model aimed at enhancing affordability and accessibility. Priced at 15 cents per million input tokens and 60 cents per million output tokens, GPT-4o mini represents a significant cost reduction compared to previous models like GPT-3.5 Turbo. The new model is more than 60 percent cheaper, making it an economical option for a wide range of applications.
GPT-4o mini is designed with a context window of 128K tokens and supports up to 16K output tokens per request. Currently, it is capable of handling text and vision tasks. Future updates will expand its capabilities to include text, image, video, and audio inputs and outputs.
The model scored 82 percent on the MMLU benchmark, surpassing GPT-4 and other competitors in both text-based reasoning and multimodal tasks. It also excels in math and coding benchmarks, outshining Gemini Flash and Claude Haiku in these areas.
In terms of safety, GPT-4o mini incorporates several measures to address risks such as harmful content and system vulnerabilities. These include content filtering and advanced techniques to mitigate issues like jailbreaks and prompt injections. GPT-4o mini is now available through various APIs and to ChatGPT users, marking a step forward in making high-performance AI more accessible.
Image credit: rarrarorro / depositphotos