ChatGPT maker OpenAI announced the launch of a new AI model, GPT-4o mini, designed to be a smaller and more affordable model.
The company revealed that it had been working to make its subscription-based AI model cheaper and faster to allow developers to build apps based on the model.
GPT-4o mini is set to be up to 60% cheaper than GPT-3.5 Turbo priced at 15 cents per million input tokens and 60 cents per million output tokens. OpenAI also says that the new AI model outperforms the GPT-4 model on chat preferences having scored 82% on Massive Multitask Language Understanding (MMLU).
MMLU is an intelligence and reasoning benchmark that is used to test the capabilities of language models. So the higher the MMLU score, the higher the level of understanding and language use, for real-world usage.
OpenAI also seems to be claiming its spot in the world of AI as they also tested the GPT-4o mini model against models like Gemini Flash by Google, which scored about 77.9%.
With this new model, OpenAI plans to also add support for image, audio and video inputs and outputs in the application interface in the future.
“GPT-4o mini enables a broad range of tasks with its low cost and latency, such as applications that chain or parallelize multiple model calls (e.g., calling multiple APIs), pass a large volume of context to the model (e.g., full code base or conversation history), or interact with customers through fast, real-time text responses (e.g., customer support chatbots),” states OpenAI.