DOOMBENCH.COM
Hourly watch
Capability gains

OpenAI releases o3-mini reasoning model across ChatGPT and API

OpenAI released o3-mini with selectable reasoning effort, function calling, structured outputs and search, extending cost-efficient STEM reasoning to free and paid ChatGPT users and API developers.

CURRENT ASSESSMENT · REVISION 1
TOWARD DOOM73confidence 97/100

Why it moved the index

The release combined near-o1 technical reasoning with production tool interfaces, lower latency and very broad access, materially increasing capable autonomous application building. Deliberative alignment, preparedness evaluations and restricted initial API tiers reduce but do not remove risk.

0 comments · 0 votesOpen discussion

Public discussion is readable by everyone. Sign in to comment, reply, or vote.

No comments yet. Start the discussion.

Send detailed or private feedback →
AUDIT TRAIL

Assessment history

  1. R1
    Toward 73 · confidence 97

    New January 2025 exact reasoning-model release with no durable model or story collision.

    12 Aug 2026