OpenAI has launched a new AI model called o1, the first model that can truly "think."

According to performance metrics, o1 surpasses the latest GPT models by a wide margin. The developers designed this neural network to solve extremely complex problems faster than a human can.

The training that underpins o1 is fundamentally different from its predecessors, utilizing an entirely new algorithm.

Practically speaking, it handles coding tasks and solves multi-step problems eight times better than previous models.

The standout feature of o1 is its ability to "reason" like a human. Its responses closely match those of individuals with advanced academic degrees.

Access to the o1-preview is available starting today for ChatGPT Plus users. You can try it out now!