OpenAI launches new powerful AI model GPT-4o with vision, audio and text capabilities
In a major announcement, OpenAI recently unveiled its latest AI model called GPT-4o that brings the advanced abilities of GPT-4 within easy reach of all users. During a webcast event, OpenAI’s CTO Mira Murati explained how GPT-4o makes the most advanced AI features more natural and simpler to use for everyone.
GPT-4o combines vision, audio and text processing into one model, allowing it to understand inputs across different modalities. Murati highlighted OpenAI’s goal of ensuring AI benefits humanity and how GPT-4o takes them closer to that mission. The new model offers faster and more intuitive interactions while addressing issues like latency faced in the past.
A key highlight was GPT-4o making all of OpenAI’s powerful tools accessible to regular users for free. Over 100 million people currently use ChatGPT and with GPT-4o, they can experience capabilities on par with paid offerings. Developers also gain opportunities with the model’s wider availability.
Live demos showcased GPT-4o’s abilities like carrying out real-time translations and understanding emotions from facial expressions. It solved math problems step-by-step via both text and drawings. Memory features and 50 language support were also presented.
Going forward, GPT-4o will see phased deployment across various uses. It processes information two times faster at half the cost compared to before. Overall, the new model is set to enhance how people access and utilize artificial intelligence through more natural interactions.
