OpenAI Launches Cost-Effective GPT-4o Mini AI Model

July 19, 2024

OpenAI has released GPT-4o mini, a new AI model that is more accessible to a wider range of people and costs less. There is a lot of competition in the AI area right now, especially from big names like Google and Meta.

What is GPT-4o mini?

The GPT-4o mini is a smaller and less expensive form of OpenAI’s other language models. It costs 15 cents per million input tokens and 60 cents per million exit tokens, which is more than 60% less than GPT-3.5 Turbo, which came before it.

Performance Metrics

OpenAI says that GPT-4o mini does better than GPT-4 in chat tastes and has an 82% score on the MMLU benchmark for Massive Multitask Language Understanding. MMLU tests how well a model can learn and use language in a variety of situations. On the other hand, Google’s Gemini Flash gets 77.9% and Anthropic’s Claude Haiku gets 73.8%. This makes the GPT-4o mini very competitive when it comes to its ability to understand words.

Availability

The mini form needs less computer power, so it can work for companies that don’t have a lot of resources. Because it is so efficient, more businesses can use creative AI in their work. ChatGPT users on the Free, Plus, and Team tiers can access GPT-4o mini instead of GPT-3.5 Turbo right away. Next week, business users will be able to do the same. The model is based on what we know now and until October 2023.

More About OpenAI

OpenAI began as a non-profit group when it was founded in December 2015. The OpenAI Gym, a set of tools for making reinforcement learning algorithms, was its first big release. In 2019, OpenAI’s GPT-2 model caused a stir because it could make text that was not what it seemed to be. OpenAI put out rules for the right way to use its AI systems in 2020. The name of the business shows that it wants to encourage open collaboration in AI research, but it changed to a capped-profit model to get the money it needs to fund its work.

« Previous Post Next Post »