OpenAI launched GPT-5, its most advanced model, described as a personal expert capable of software creation on demand. Compared to previous models, GPT-5 shows enhancements in coding, writing, math, and visual perception, with significant reductions in hallucinations and deceptive behavior. It consists of several models routed based on user intent and task complexity. OpenAI plans to integrate these into a single model in the future. The architecture increases efficiency and reduces output tokens while maintaining high performance across various capabilities. Paid users can also enable reasoning functionality permanently.
OpenAI says GPT-5 delivers improvements in coding, writing, math, and visual perception, while also cutting down on hallucinations and deceptive behavior.
This routing model is apparently being continuously trained on new input signals to make it smarter about which model it routes the request to.
GPT-5 gets more value out of less thinking time, performing better than OpenAI o3 with 50-80 percent less output tokens.
Paid users will also have the option of toggling on reasoning functionality permanently if desired.
Collection
[
|
...
]