OpenAI

gpt-4.1-2025-04-14

Made by OpenAI. Released 2025-04-14. Accessibility not recorded. Cheapest standard rate we hold: $12 per million output tokens, no context split.

What it costs

TierKindInOutCached inFrom
batchtext$1.50$6$0.50seller
standardtext$3$12$0.75seller

per million tokens, US dollars. * marks a rate from a single registry or two that disagree — the best available number rather than a confirmed one. Every other row is confirmed against a seller or a licensed index.

What it scores, by house

BenchmarkHouseScoreEvaluatedScaffold
chess puzzlesEpoch AI0.1±0.022026-08-07
frontiermathEpoch AI0.1±0.012025-04-14
frontiermath tier 4Epoch AI0.0±0.002025-07-01
frontiermath tiers 1 3 v2Epoch AI0.1±0.012026-08-27
gpqa diamondEpoch AI0.7±0.032025-04-14
math level 5near its ceilingEpoch AI0.8±0.012025-04-14
otis mock aime 2024 2025near its ceilingEpoch AI0.4±0.062025-04-14
simpleqa verifiedEpoch AI0.3±0.012026-08-31
swe bench verifiedEpoch AI0.5±0.022026-02-08
aider polyglotcollected by Epoch AI52.4
ale benchcollected by Epoch AI558
arc aginear its ceilingcollected by Epoch AI0.1
arc agi 2collected by Epoch AI0.0
cad evalcollected by Epoch AI0.4
dtbenchnear its ceilingcollected by Epoch AI0.7
enigma evalcollected by Epoch AI0.0
fictionlivebenchcollected by Epoch AI0.6
forecastbenchcollected by Epoch AI61.5
geobenchcollected by Epoch AI3791
hlecollected by Epoch AI0.1
lmcacollected by Epoch AI25.6
simplebenchcollected by Epoch AI0.3
weirdmlcollected by Epoch AI0.4

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.

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