OpenAI has launched its most capable artificial intelligence model to date, showcasing abilities in complex reasoning, scientific analysis, and real-time problem-solving that have stunned researchers both inside and outside the company. The new model is available immediately to subscribers of the premium tier.
In independent benchmarks conducted by academic institutions, the model outperformed previous state-of-the-art systems on every major evaluation metric, including tasks requiring extended chains of logical reasoning that previous models consistently struggled with.
“What we're seeing is not just incremental improvement. The gap between this model and previous generations on complex reasoning tasks is qualitatively different.”
The release intensifies competition with rival AI laboratories from Google, Anthropic, and a growing number of open-source projects that are rapidly closing the capability gap with commercial systems.