An updated version of the GPT-4 model, which powers OpenAI's flagship product, ChatGPT, is being introduced as GPT-4o. In a livestream announcement on Monday, OpenAI CTO Mira Murati stated that the revised model "is much faster" and enhances "capabilities across text, vision, and audio." Everyone can use it for free, and paying users will still be able to "have up to five times the capacity limits" compared to free users, according to Murati.

OpenAI states in a blog post that while GPT-4o's features "will be rolled out iteratively," its text and image capabilities will be available in ChatGPT.

According to a post by OpenAI CEO Sam Altman, the model is "natively multimodal," meaning it can produce content and comprehend speech, text, or image directions. According to Altman on X, the API for GPT-4o is twice as fast and half as expensive as GPT-4 Turbo, so developers who wish to play around with it can do so.

What is GPT-4o?

Open AI’s most recent flagship model, GPT-4o, has GPT-4-level intelligence but operates far more quickly and has enhanced text, voice, and vision capabilities.

GPT-4o is far more adept at comprehending and debating the photographs you share than any other model. For instance, you can now ask GPT-4o to interpret a menu written in a language other than your own, as well as provide recommendations and information about the significance and history of the dish. Future developments will enable more organic, real-time speech conversations as well as real-time video conversations using ChatGPT. You could display a live sporting event on ChatGPT and ask it to walk you through the rules. In the upcoming weeks, we want to roll out a new Voice Mode with these new features in an alpha version, giving Plus subscribers early access before going wider.

GPT-4o's language capabilities have enhanced the quality and speed of advanced AI to make it more widely available and beneficial. ChatGPT (opens in a new window) now supports more than 50 languages for user settings, sign-up and login, and other features.

GPT-4o is now available to ChatGPT Plus and Team users, and Enterprise users will soon be able to access it. Today, we are also beginning to roll out ChatGPT Free with usage caps. The message limit for Plus members is up to 5x more extensive than that of free users, and it is considerably higher for Team and Enterprise users.

Making use of screenshots and video

Video is now another way to communicate with ChatGPT. This allows you to share a real-time video of an issue you're having, like a math problem, and get assistance from other users. ChatGPT will either provide you with the solution or assist you in solving it independently.

Along with asking ChatGPT about previous talks, searching for real-time information within a conversation, and performing complex data analysis by uploading charts or code before asking questions, you can also share screenshots, photos, and documents containing text and graphics.

GPT-4, released in March 2023, was previously accessible for $20 monthly through the ChatGPT Plus subscription. It employs one trillion parameters, or bits of data, to answer inquiries. GPT-3.5, an even earlier version with a lower context window of 175 billion parameters, was freely available.

"We are very interested in the next frontier," Murati said. "So soon, we'll update you on our progress towards the next big thing."

In summary

With the release of GPT-4o, OpenAI has significantly advanced artificial intelligence. By utilizing a more rapid and adaptable paradigm, users can anticipate improved interactions with text, voice, and image inputs. Real-time voice discussions and live video engagements with ChatGPT are two examples of OpenAI's commitment to developing more engaging and user-friendly AI experiences.

Furthermore, OpenAI's dedication to democratizing cutting-edge AI technology is demonstrated by the launch of GPT-4o to multiple user tiers, ranging from ChatGPT Plus to Enterprise, and the expansion of language support to more than 50 languages. GPT-4o creates the foundation for future advancements that will further expand the potential of AI-driven communication and problem-solving by facilitating more natural interactions.