Benchmark Almanac

codex-mini-2025-05-16

codex-mini-2025-05-16.

Released May 16th, 2025 Compare with others
4
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Unknown lab May 2025

codex-mini-2025-05-16

No Epoch Capabilities Index score yet
  • ProxyWar 31.2% #7 /17
  • ARC-AGI-2 1.27% #84 /101
  • ARC-AGI 27.3% #81 /98

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Math and science

BenchmarkScoreRankSource
ARC-AGI 27.3% #81 of 98 Official leaderboard
ARC-AGI-2 1.27% #84 of 101 Official leaderboard

Agents and tools

BenchmarkScoreRankSource
ProxyWar 31.2% #7 of 17 Official leaderboard

Safety and security

BenchmarkScoreRankSource
A.S.E (AICGSecEval) 29.7 #16 of 21 Official leaderboard

What does it cost to run?

On the 2 leaderboards that publish a price, its best run usually costs 64% of the median model's. On every one, another model comes close to its score for less.

Benchmark Score Price From cheapest to priciest Against the median Close to its score for less Close for less
ARC-AGI-2 1.27% $0.23 0.5× the median 0.5× GPT-5 nano 0% for $0.0025
ARC-AGI 27.3% $0.16 0.8× the median 0.8× GPT-6 Luna 37.7% for $0.0032

Who measured it?

Independent runs cover all 4 of its benchmarks.

  • An independent run and a claim 1
  • An independent run only 3
Official leaderboards scored it on 4.

Sources: Official leaderboards, Epoch AI (CC BY 4.0) and labs’ model cards