AI Model Timeline

Tracking the accelerating release frequency of frontier AI models.

o4-mini

OpenAI·

LLMsreasoningcloud

Overview

Fast, cost-effective reasoning model released alongside o3.

Capabilities and innovations

Fast reasoningCost-effectiveTool use

Benchmarks

BenchmarkScoreSource
GPQA Diamond76.8%unsourced
SWE-bench Verified68.1%unsourced
MMLU89.5%unsourced
  • Humanity’s Last Exam (no tools): not reported by the vendor
  • 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.