OpenAI Launches GPT-6 With Real-Time Internet Access and Multimodal Reasoning
OpenAI's GPT-6 arrives with live internet access, multimodal reasoning, and novel-length context windows, outperforming rivals on 12 of 15 benchmarks.
Image: GlobalBeat / 2026
OpenAI unveiled its most advanced language model to date on Thursday, introducing GPT-6 with native real-time internet access, enhanced multimodal reasoning, and a dramatically expanded context window that the company says makes it capable of processing the equivalent of a full novel in a single query.
The model demonstrates significant improvements in mathematical reasoning, code generation, and multi-step logical inference compared to its predecessor. In benchmark testing, GPT-6 outperformed every previously published model on 12 of 15 standard evaluation criteria.
“This is not an incremental update,” said OpenAI’s Chief Executive during the launch event in San Francisco. “We believe GPT-6 represents a fundamental shift in what AI can do reliably.”
The rollout begins with ChatGPT Plus subscribers before expanding to the API. Pricing for enterprise customers has not yet been disclosed.
The announcement prompted swift responses from Google and Anthropic, both of which hinted at upcoming model releases. Shares of Microsoft, OpenAI’s largest backer, rose 2.1 percent on the news.