OpenAI Chief Technology Officer Mira Murati recently revealed that the company’s next-generation artificial intelligence model, GPT-5, is expected to be launched by the end of 2025 or early 2026. The model is expected to achieve PhD-level intelligence in specific tasks, marking a major leap forward in artificial intelligence technology.

GPT-5, internally codenamed "Gobi" and "Arakis," is a multimodal model with 520 trillion parameters, a significant increase from the approximately 20 trillion parameters of the previous generation GPT-4. Such a large parameter scale shows its potential power. Murati compared the progress from GPT-4 to GPT-5 to the leap from high school to college, indicating a significant increase in the sophistication and capabilities of the new model.

It is worth noting that the timing of this release is very different from early rumors. The market had previously predicted that GPT-5 might be released by the end of 2023 or the summer of 2024. However, Mulati’s latest statement pushed the time to late 2025 or early 2026. OpenAI CEO Sam Altman emphasized that there is no set release schedule yet.