OpenAI Launches GPT-5.4 Thinking for Professional Reasoning

OpenAI officially released GPT-5.4 Thinking to enhance professional reasoning and multi-step workflows ^[1][2]. This new model features a massive 1-million-token context window, allowing for deeper analysis of large datasets ^[1][2]. A key innovation is native computer-use capability, which enables the model to interact with software applications directly ^[1]. OpenAI also introduced a steerability feature that lets users adjust the AI's reasoning plan while it generates a response ^[2][3]. These tools aim to improve productivity in specialized fields like coding, analytics, and complex research ^[1]. Industry observers noted that the model's speed and reasoning capabilities set a new benchmark for competitive AI performance ^[3].