The update, aimed at improving overall user experience on ChatGPT, comes alongside a move to double the hourly usage limits for GPT-4o and GPT-4-mini-high models for ChatGPT Plus subscribers, following feedback from high-usage customers.
Altman also confirmed that from April 30, the GPT-4 model will be removed from ChatGPT as the platform transitions fully to GPT-4o, now the default model for all users.
GPT-4o integrates text, image, and audio capabilities, offering faster response times and a more seamless user experience, as the company works to consolidate its AI stack around a single, multimodal model.
Discover the stories of your interest
The latest enhancements come at a time when OpenAI continues to face infrastructure constraints, particularly a shortage of graphics processing units (GPUs). Altman has acknowledged the challenge, saying the company is making “hard trade-offs” between introducing new features and maintaining performance.
To address the growing demand, OpenAI is expanding its compute infrastructure by adding tens of thousands of GPUs.
In addition to the GPT-4o update, OpenAI has made its image-generation model, gpt-image-1, available to developers and businesses globally via API, further extending its multimodal capabilities across platforms.
Organisations can now integrate the API into their products to generate high-quality images from text prompts, enabling use cases across design, e-commerce, marketing, education, and gaming.