AI Model Timeline

Tracking the accelerating release frequency of frontier AI models.

o3-mini

OpenAI·

LLMsreasoningcloud

Overview

Smaller, faster reasoning model bringing o-series capabilities to a more accessible form.

Capabilities and innovations

ReasoningFast inferenceCost-effectiveCompact ReasoningEfficient CoTAccessible Intelligence

Benchmarks

BenchmarkScoreSource
GPQA Diamond79.7%unsourced
Humanity’s Last Exam (no tools)13.4%unsourced
SWE-bench Verified49.3%unsourced
MMLU86.7%unsourced
  • SWE-bench Pro: benchmark did not exist at release
  • MMLU-Pro: not reported by the vendor
  • Terminal-Bench 2.x: benchmark did not exist at release
  • MMMU-Pro: benchmark did not exist at release

Links

More from OpenAI

Data curated by AI Model Timeline. See the methodology for admission criteria, benchmark eras and source priority.