OpenAI has launched GPT-4o (“o” for “omni”), a new version of its AI model, and a dedicated desktop application, promising a more efficient and integrated experience for users. GPT-4o is designed to operate faster than its predecessor, GPT-4, and aims to make user interaction more natural.
In that context, during the event, the OpenAI team offered a live demo of the new prowess of the GPT-4o with live conversations. The conversations highlighted the quick response time of the new GPT-4o-powered ChatGPT. OpenAI claims it can respond to audio inputs in as little as 232 milliseconds, with an average of 320 milliseconds, which is similar to human response time.
It is capable of processing text, vision, and audio across different formats. The desktop app is designed to effortlessly integrate into user workflows. New features include the ability to upload documents and screenshots, use memory for continuity in conversations, and browse information directly during interactions.
The AI model now supports improved performance in 50 languages and can analyse charts. GPT-4o is available not only for personal and professional use but also offers enhanced developer APIs. The enhancements include a performance boost making the model twice as fast and a cost reduction, making it 50 per cent cheaper to deploy.
OpenAI in a blogpost explains the new feature update saying, "GPT-4o is a step towards much more natural human-computer interaction—it accepts as input any combination of text, audio, and image and generates any combination of text, audio, and image outputs."
Paid vs Free version
Paid users will continue to have five times the capacity limit compared to standard offerings. The rollout of GPT-4o will be gradual, with deployments planned over the next few weeks, confirmed OpenAI CTO Mira Murati. OpenAI claims its teams are working to ensure a smooth transition and integration for all users.