/dq/media/media_files/CctHSwd8Vq2YBPsl4bBp.jpg)
OpenAI has released GPT-5, a new version of the AI technology behind its ChatGPT chatbot. This launch follows nearly two years after the release of GPT-4. The new model is available to all ChatGPT users, with some differences in access and usage limits based on their subscription plans.
GPT-5 is designed to function as a unified system, combining the strengths of previous models. It uses a real-time router to determine whether a query requires a quick answer or a more detailed, reasoned response. This system is trained on real-world feedback to ensure each response is as helpful as possible.
OpenAI describes GPT-5 as a step forward in performance and intelligence. The company says the model shows state-of-the-art performance in key areas. For example, it is more coherent in writing and more accurate in math, offering step-by-step reasoning for complex problems. The model also shows improved understanding and safety in health-related queries. It demonstrates stronger image interpretation and description, and follows instructions more precisely. GPT-5 also addresses persistent issues seen in earlier models, reducing hallucinations and improving factual consistency.
The coding capabilities have also been enhanced. OpenAI presents GPT-5 as its strongest coding model yet, excelling at generating front-end code, debugging large codebases, and design-focused development. The company states that developers using GPT-5 have seen major improvements in handling both logic and design, which helps reduce iteration time.
OpenAI CEO Sam Altman described the new model as a significant step toward artificial general intelligence. He said the experience is like talking to a "PhD-level expert in anything." OpenAI is making the model available to everyone. Free-tier users can use GPT-5 with usage limits. Paid subscribers get higher limits, and Pro subscribers receive GPT-5 Pro, a specialised version with extended reasoning abilities.
The launch of GPT-5 introduces some new features for ChatGPT users. There are four new response styles, Cynic, Robot, Listener, and Nerd, that change how the assistant replies without specific instructions. Additionally, the improved Voice Mode now adapts its responses and is available in custom GPTs.