OpenAI, which has been shaping the direction of the artificial intelligence industry with its recent strides, is preparing to unveil its next-generation language model, GPT-5, in August. Details about the model, which the company has not yet officially announced, have emerged from sources familiar with the matter. Unlike previous versions, GPT-5 is expected to stand out with its multi-model architecture.

OpenAI to unveil GPT-5 in August

According to information reported by The Verge, GPT-5 is being developed with a collaborative architecture of multiple special-purpose models rather than a single AI engine. This new architecture aims to enable the system to respond more adaptably and effectively to different types of tasks. The approach relies on combining GPT and the “o” series of models under a single framework. This architecture could play a key role in the transition to multi-purpose AI systems by increasing flexibility and modularity.

OpenAI CEO Sam Altman stated in February that the company is working on a hybrid platform that combines the o3 model with GPT-based systems. GPT-5 is expected to be the first concrete example of this strategy. It’s stated that the new architecture will go beyond text generation and incorporate multiple tasks such as coding, visual analysis, and logical reasoning under one roof.

However, development may not proceed as planned. Internal sources indicate that some technical challenges and server capacity bottlenecks are still being resolved. It’s also stated that the successive announcements of new models in the market may also influence the release date.

Details regarding the user levels to which the model will be available, and whether it will be rolled out gradually as with GPT-4 or directly with broad access, have not yet been finalized. However, GPT-5 is expected to be integrated into OpenAI’s products such as Copilot, ChatGPT, and custom API services.