GPT-4o debuts @OpenAI Spring 2024; paradigm to future collaboration

With GPT-4o, OpenAI is launching its newest flagship model and making more capabilities, across text, video, and audio. In future, it will be looking at interactions between ourselves and machines, shifting paradigm to future of collaboration.

Pradeep Chakraborty
New Update

Mira Murati introduces GPT-4o.

At the OpenAI Spring update 2024, the company has introduced GPT-4o and more tools to ChatGPT free users.


Mira Murati, CTO, OpenAI, introduced the GPT-4o and the company's mission. It includes desktop app and Web UI update, GPT-4o, GPT-4o free access, etc. We are ensuring AI general benefits to all of the humanity. We have also upgraded the UI. We want the user experience to be more natural. We want users to focus on the collaboration with ChatGPT. 

We recently made ChatGPT available without the signup flow. We want you to use this wherever you are. We want the interaction to become more natural, easy for everyone.

We are now releasing the new flagship model -- GPT-4o. We are launching our newest flagship model and making more capabilities, across text, video, and audio. It is available for free in ChatGPT. It is much faster. It brings GPT 4-level intelligence to everyone. 


This is the first time we have made huge progress towards ease of use. In the future, we will be looking at interactions between ourselves and the machines. GPT-4o is really shifting the paradigm to the future of collaboration. 

With voice mode, we have three models coming together to deliver this experience. These are transcription, intelligence, and text-to-speech. This also brings lot of latency to the experience. With GPT-4o, it all happens natively. GPT-4o reasons across voice, text, and vision. It also allows us to bring Jupiter IV intelligence to free users. 

Today, we have over 100 million people learning, creating, and working on ChatGPT. We have advanced tools only available to our paid users up until now. With GPT-4o, we can now bring these tools to everyone. Starting today, you can make use of thousands of apps from the GPT store. Now, it will have much bigger audience. This will include university professors, podcasters, etc.


You can also use vision. You can upload screenshots, photos, documents with text and images, etc., and start conversations with ChatGPT about all this content. You can also use the memory. It makes GPT more useful and helpful. You can remember things you discuss for future chats. You can do advanced data analytics, and also analyze data and create charts. 

We have also improved on the quality and speed in 50+ different languages. That covers at least 97% of the world's Internet population. We want to bring this experience to as many people as possible. GPT-4o gives you free access, along with GPT store, with vision, browsing, memory, and advanced data analytics. 

We are also bringing GPT-4o to API. It is 2x faster, 50% cheaper, and 5x higher rate limits compared to GPT-4 Turbo. Our developers can start building today with GPT-4o, and make amazing AI apps, deploying them at scale. GPT-4o brings new challenges to us. Our teams are working to mitigate against misuse. 

Later, Mark Chen and Barret Zoph displayed some demos.

openai GPT-4o