OpenAI officially launched GPT-5 today, its most advanced large language model to date, featuring significant improvements in multimodal reasoning, mathematical problem-solving, and the ability to learn from user interactions within a single session. The model is available immediately through the ChatGPT Plus subscription and the OpenAI API.
GPT-5 demonstrates a 40% improvement over GPT-4o on graduate-level reasoning benchmarks and can natively process text, images, audio, and video inputs in a unified context window. Perhaps most notably, the model features what OpenAI calls adaptive context, the ability to refine its responses based on user corrections and preferences during a conversation without requiring fine-tuning.
OpenAI CEO Sam Altman described GPT-5 as a significant step toward artificial general intelligence while acknowledging that the model still has limitations, including occasional confabulations and difficulty with highly specialized domain knowledge. Pricing for API access starts at $10 per million input tokens and $30 per million output tokens. Competitors including Anthropic, Google, and Meta are expected to announce their own next-generation models in the coming weeks.
