OpenAI is set to launch its next-generation AI model, GPT-5, as early as August. The company’s CEO, Sam Altman, had first outlined a roadmap for the model five months ago. According to sources, Microsoft engineers began preparing server capacity for GPT-5 in late May but faced testing and development challenges that pushed the timeline back.
Altman recently demonstrated GPT-5’s capabilities on Theo Von’s podcast, showcasing its ability to answer questions he couldn’t. The AI model has been highly anticipated since the launch of GPT-4 in March 2023. However, it seems the company saved the “GPT-5” name for a future release.
When launched, GPT-5 will come with “mini” and “nano” versions available through its API. The main version, combining a large language model and simulated reasoning, will be accessible via ChatGPT and OpenAI’s API. In contrast, the nano version will only be accessible via the API.
Source: https://arstechnica.com/ai/2025/07/openais-most-capable-ai-model-gpt-5-may-be-coming-in-august