o1-mini

Retired · Language, Reasoning

o1-mini is a smaller, lower-cost OpenAI reasoning model released on 12 September 2024 together with o1-preview and aimed at STEM tasks, especially math and coding. It was offered to API usage tier 5 developers and in paid ChatGPT plans; OpenAI says its knowledge of non-STEM facts is comparable to small models like GPT-4o mini. [1] Primary source

Timeline of o1-mini →

Claims and evidence

  • Released Primary source[1]second paragraph [2]Sep 12, 2024
  • Deprecated Primary source[3]2025-04-28: o1-preview and o1-mini
  • Retired Primary source[3]2025-04-28: o1-preview and o1-mini
  • Status Retired Primary source[3]2025-04-28: o1-preview and o1-mini
  • Replaced by o4-miniRecommended replacement listed on the deprecations page when checked on 2026-10-01; the o1-mini model page recommends o3-mini instead. Primary source[3]2025-04-28: o1-preview and o1-mini
    • Change · Efficiency Faster than o1-preview and 80 percent cheaper in the API, with higher rate limits in ChatGPT.Compared with o1-preview Primary source[1]
    • Change · Other Narrower knowledge outside STEM, such as dates, biographies and trivia, which OpenAI compares to small models like GPT-4o mini.Compared with o1-preview Primary source[1]
    • Input text Primary source[4]Model details
    • Output text Primary source[4]Model details
    • Context window 128,000 tokens Primary source[4]Model details
    • Parameters Not disclosedNot disclosed by OpenAI; the announcement only calls it a smaller model. No claim madeNo source
    • Access API, consumer appAPI usage tier 5 at launch; ChatGPT Plus, Team, Enterprise and Edu. Primary source[1]second paragraph

    Lineage

    Predecessors

    No known predecessor.

    Successors

    Based on

    Not derived from another model.

    Variants and derived

    None recorded.

    Siblings

    None recorded.

    All ancestors

    None.

    All descendants

    Variants

    No variants recorded in this record.

    Related AI Radar coverage

    AI Radar coverage starts in June 2026; no coverage linked yet.

    All model releases from OpenAI on AI Radar →