OpenAI ships GPT-4.1 with a million-token window

In April 2025, OpenAI launched GPT-4.1 alongside cheaper mini and nano models, available through the developer API. The headline changes: stronger coding and instruction-following, plus a one-million-token context window that lets the model read far more of a document, codebase, or transcript in a single pass.

Pricing dropped too, with a wide spread between the flagship and the nano tier. That means you can route simple tasks to a cheap model and save the expensive one for work that needs it.

For a small or mid-size business, the practical takeaway isn't the benchmark scores. It's that summarizing long PDFs, tagging support tickets, or drafting first-pass copy now costs less and handles bigger inputs. Start with one narrow, boring task, measure whether it saves time, then expand. The tiered pricing rewards matching the model to the job rather than reaching for the biggest one out of habit.