GPT-4.1 mini
OpenAI · United States · 2025
A million-token window for forty cents per million input tokens — the cheapest long context OpenAI had sold up to that point.
GPT-4.1 mini launched on 14 April 2025 as the middle tier of the GPT-4.1 family. It keeps everything that defined the generation — the 1,047,576-token context window, 32,768-token output limit, text and image input, and a knowledge cutoff of 1 June 2024 — and cuts the price by four fifths. Input costs $0.4 per million tokens, cached input $0.1, and output $1.6. That combination mattered more than any benchmark: it made document-scale prompts affordable for routine work, where before a million-token window had been reserved for models priced like flagships. Like the base model, it has no reasoning step and is aimed at instruction following and tool calling; OpenAI points developers to GPT-5 mini for harder problems. It was left off the April 2026 shutdown list and, as of August 2026, is still served by the API — unlike GPT-4.1 nano below it, which switches off on 23 October 2026.
▸Videos
No videos yet.
▸Reviews
No reviews yet. Be the first!