Models will commoditize. Capabilities will converge. What will endure are the interfaces agents already rely on, and the data and execution capabilities behind them.
GPT-5.4 is also more reliable, producing 18% fewer errors and 33% fewer false claims than GPT-5.2, according to OpenAI.
OpenAI launches GPT-5.4, calling it its most capable and efficient AI model yet, with AI agents, computer control, improved reasoning, and a 1M-token context.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results