OpenAI has launched a research preview of its latest AI model, GPT-4.5, for ChatGPT Pro users and developers worldwide. The company plans to extend access to ChatGPT Plus and Team users next week, followed by Enterprise and Edu users.
According to OpenAI, GPT-4.5 enhances pattern recognition, improves creative insights, and demonstrates greater emotional intelligence. The model also significantly reduces hallucination rates—a common issue where AI generates false information. At 37.1%, GPT-4.5’s hallucination rate is lower than GPT-4o’s 61.8% and its o1 reasoning model’s 44%.
CEO Sam Altman described GPT-4.5 as “a giant, expensive model,” revealing that GPU shortages have hindered a full-scale rollout. “We will add tens of thousands of GPUs next week,” he posted on X, noting the challenge of predicting growth surges that cause hardware shortages.
Despite its advancements, GPT-4.5 does not currently support voice and video capabilities. However, it does allow file and image uploads and is optimized for writing and coding projects.
This release reinforces OpenAI’s leadership in AI innovation, following its rapid expansion with models like GPT-4o. As AI competition intensifies, OpenAI’s latest upgrade aims to deliver more powerful and reliable tools for users worldwide.
