OpenAI

o3-mini-2025-01-31

Made by OpenAI. Released 2025-01-31. Accessibility not recorded. We hold no price for this model: no seller we can read lists it any more, so what follows is what was measured, not what it cost.

What it scores, by house

BenchmarkHouseScoreEvaluatedScaffold
chess puzzlesEpoch AI0.1±0.032026-08-07medium
chess puzzlesEpoch AI0.1±0.022026-07-15low
chess puzzlesEpoch AI0.2±0.042025-12-08high
frontiermathEpoch AI0.1±0.022025-11-16high
frontiermathEpoch AI0.1±0.012025-03-06medium
frontiermath tier 4Epoch AI0.0±0.032025-07-01high
frontiermath tier 4 v2Epoch AI0.0±0.002026-06-11high
frontiermath tiers 1 3 v2Epoch AI0.1±0.022026-08-27medium
frontiermath tiers 1 3 v2Epoch AI0.0±0.012026-08-27low
frontiermath tiers 1 3 v2Epoch AI0.2±0.022026-06-11high
gpqa diamondEpoch AI0.7±0.032026-07-20low
gpqa diamondEpoch AI0.8±0.032025-02-13high
gpqa diamondEpoch AI0.7±0.032025-01-31medium
math level 5near its ceilingEpoch AI1.0±0.002025-02-13high
math level 5near its ceilingEpoch AI1.0±0.002025-01-31medium
mystery game puzzlesEpoch AI0.1±0.032026-08-27high
otis mock aime 2024 2025near its ceilingEpoch AI0.4±0.072026-07-20low
otis mock aime 2024 2025near its ceilingEpoch AI0.8±0.052025-02-27high
otis mock aime 2024 2025near its ceilingEpoch AI0.6±0.062025-02-25medium
simpleqa verifiedEpoch AI0.2±0.012026-08-31high
aider polyglotcollected by Epoch AI53.8medium
aider polyglotcollected by Epoch AI60.4high
arc aginear its ceilingcollected by Epoch AI0.1low
arc aginear its ceilingcollected by Epoch AI0.3high
arc aginear its ceilingcollected by Epoch AI0.2medium
arc agi 2collected by Epoch AI0.0medium
arc agi 2collected by Epoch AI0.0low
arc agi 2collected by Epoch AI0.0high
cad evalcollected by Epoch AI0.5medium
critptcollected by Epoch AI0.0high
cybenchcollected by Epoch AI0.2medium
dtbenchnear its ceilingcollected by Epoch AI0.7high
fictionlivebenchcollected by Epoch AI0.4medium
forecastbenchcollected by Epoch AI59.6unknown
gsocollected by Epoch AI0.0OpenHands
lech mazur writingcollected by Epoch AI6.2high
lech mazur writingcollected by Epoch AI6.2medium
live benchcollected by Epoch AI75.9high
live benchcollected by Epoch AI62.5low
live benchcollected by Epoch AI70.0medium
lmcacollected by Epoch AI19.0high
scicodecollected by Epoch AI0.4high
simplebenchcollected by Epoch AI0.2high
weirdmlcollected by Epoch AI0.4high

These scores are not comparable between houses and are not added up. A score is dated by when it was evaluated, not when the model was released, and a score without its scaffold is not a measurement — which is why both are printed.

What retires, and when

  • deprecation2026-10-23

Every rate we hold · Why there is no single best model ·