OpenAI on Thursday released a smaller and more affordable version of its flagship large language model that powers ChatGPT. The new model, called GPT-4o mini, will cost developers 60 percent less to build AI-powered apps and services with as compared to GPT-3.5 Turbo, Open’s smallest model until today. But the big news here is for consumers. GPT-4o mini will replace GPT-3.5 Turbo for free users of ChatGPT starting today — which means that your baseline ChatGPT experience will improve significantly.
OpenAI claimed that GPT-4o mini achieved an 82 percent score on an industry benchmark called the MMLU, which stands for Measuring Massive Multitask Language Understanding, and includes 16,000 multiple-choice questions across 57 academic subjects. That’s slightly lower than GPT-4o, which scored 88.7 percent, but higher than GPT-3.5 Turbo, which scored just 70 percent. AI experts have warned against relying on these kinds of benchmarks to measure how smart AI systems are, but so far, they’re the only way to measure the performance of large language models.
Smaller versions of flagship models give developers more flexibility when it comes to building AI-powered apps. After all, not everyone needs or can afford access to the largest, most powerful models for every use case. In May, Google announced Gemini 1.5 Flash, its own lightweight model that the company said was optimized for speed and efficiency. Other AI companies like Anthropic, too, have smaller versions of their full-scale models.
GPT-4o mini can currently take in and generate text and images, but the model will eventually be able to process other types of content like audio and video. And even though GPT-3.5 Turbo is going away from ChatGPT, developers can still access it via OpenAI’s API to build apps and services with it for now — until it eventually goes away too.