OpenAI o1-pro

o1-pro reached the API on 19 March 2025, as one of the first models offered exclusively through the new Responses API. The idea was simple and blunt: take o1 and give it more compute per question, so that it thinks harder and answers more consistently. The price was the headline. o1-pro listed at 150 dollars per million input tokens and 600 per million output — ten times plain o1, and to this day the highest list price OpenAI has published for any model. There was no cached-input discount at all, so the usual trick of amortising a long prompt across many calls did not apply. Rate limits were correspondingly tight: 30,000 tokens per minute at the entry tier. Technically it matched the later o-series: a 200,000-token context window, up to 100,000 output tokens, text and image input, text output, and a knowledge cutoff of October 2023 inherited from o1. Chat Completions was not supported. OpenAI announced the deprecation on 22 April 2026. The o1-pro-2025-03-19 snapshot shuts down on 23 October 2026, and the recommended replacement is gpt-5.6-sol with reasoning mode set to pro — the same shift seen across the line, where a separate premium model becomes a parameter on a general one.

▸Videos

No videos yet.

▸Reviews