GPT-4.5 Turbo Launches with 256K Context Window and 40% API Price Cut

OpenAI released GPT-4.5 Turbo, expanding the context window to 256K tokens and reducing API pricing by approximately 40% versus GPT-4o. The model shows improved instruction following on long-document tasks and is available immediately via the OpenAI API and Azure OpenAI Service. Batch API support is included at launch, enabling asynchronous processing at an additional 50% discount.