What is the IQ of o1?

o1 has an estimated IQ of 95, which ranks #148 out of 176 public models on the AI IQ leaderboard. The score combines seven equally weighted reasoning dimensions using measured benchmark results and conservative estimates for missing coverage.

Among the dimensions with published benchmark results, o1 ranks highest in Reliability (IQ 110, #108 of 177) and Programmatic Reasoning (IQ 101, #120 of 177). Its lowest-ranked measured dimension is Computer Use (IQ 80, #144 of 176).

What is the EQ of o1?

o1 doesn't have an EQ score yet. EQ (emotional reasoning) is calculated from EQ-Bench 4 and AttuneBench, and neither has published a result for o1. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.

Which benchmarks has o1 been tested on?

o1 has published results on 16 of the 39 benchmarks behind AI IQ's Composite IQ, including MathArena, IFBench, MMLU-Pro, SimpleQA Verified and AA Omniscience.

How much does o1 cost?

o1 costs $15.00 per million input tokens and $60.00 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $74.68 per million input-plus-output tokens.

No — o1 is a proprietary model. Its weights are not publicly available, so it is used through hosted APIs and products.

o1 was developed by OpenAI, based in the United States.

When was o1 released?

o1 was publicly released on December 5, 2024.

How large is the context window of o1?

AI IQ hasn't verified an input context limit for o1 from its developer's documentation, so none is listed rather than guessing.

AI IQ doesn't have independent speed measurements for o1 yet. Speed comes from independent tests of a model's hosted API, which aren't available for every model — especially new, restricted-access, or self-hosted ones.

OpenAIReleased Dec 5, 2024

Proprietary

o1

Its best results are in Reliability (#108 of 177) and Programmatic Reasoning (#120 of 177); weakest in Computer Use (#144 of 176). Knowledge Work is estimated, because no results are published yet.

Composite IQ ?

#148 of 176 ranked models · ahead of 16%

OpenAI's #20 model, behind gpt-6.1-sol (141)

Effective cost

$74.68/1M

$15 in · $60 out

Speed

Not measured

No speed data yet

Context

Unverified

Not in publisher docs

Evidence

6of 7 dims

16 of 39 benchmarks

Compare with gpt-6.1-solAPI pricing

IQ profile

Seven dimensions, averaged equally into Composite IQ

Compare shape with

MeasuredEstimated95% range

#1 in dimensionTop-10 median

IQ dimensions

Ranks are among models scored in each dimension. Ranges are model-based estimates, not calibrated confidence intervals.

DimensionIQPosition · 70–160RankEvidencevs Top 10

Abstract ReasoningAbstract

79measured

#163 / 201

2/5

-47.5

Mathematical ReasoningMath

102measured

#120 / 176

3/5

-45

Academic ReasoningAcademic

106measured

#149 / 203

5/6

-29.5

Programmatic ReasoningCoding

101measured

#120 / 177

1/6

-48.5

Computer UseComputer Use

80measured

#144 / 176

1/5

-63

ReliabilityReliability

110measured

#108 / 177

4/6

-24

Knowledge WorkKnowledge Work

90estimated

#167 / 176

0/6

-41.5

o195% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human

Benchmark results

16 of 39 Composite IQ benchmarks have a published result for o1. Benchmarks without a result stay listed so gaps are visible.

Abstract Reasoning IQ 79 · measuredBest: #60 of 76 on Chess Puzzles2/5

BenchmarkScorePercentileRankLeader · source

Chess Puzzles

Epoch AI vEpoch 100-position leaderboard (task implementation versions recorded per run) · Oct 14

15.0%

#60 / 76

Leader gpt-6-astra · 72.0%
Epoch AI vEpoch 100-position leaderboard (task implementation versions recorded per run) · Oct 14

ARC-AGI-1

ARC Prize verified leaderboard · Sep 22

21.0%

#61 / 68

Leader fable-5 · 98.5%
ARC Prize verified leaderboard · Sep 22

Not yet reported

ARC-AGI-3ARC-AGI-2Mystery Game Puzzles

Mathematical Reasoning IQ 102 · measuredBest: #29 of 64 on MathArena3/5

BenchmarkScorePercentileRankLeader · source

MathArena

MathArena Models · Sep 22

31.4%

#29 / 64

Leader gpt-6.1-sol · 84.0%
MathArena Models · Sep 22

AIME

71.5%

#45 / 56

Leader gemini-3.1-pro · 98.1%

FrontierMath Tier 1-3

Full source leaderboard reconciliation · Oct 23

14.7%

#77 / 84

Leader gpt-6-astra · 93.7%
Full source leaderboard reconciliation · Oct 23

Not yet reported

FrontierMath Tier 4ProofBench 1.1

Academic Reasoning IQ 106 · measuredBest: #24 of 45 on MMLU-Pro5/6

BenchmarkScorePercentileRankLeader · source

MMLU-Pro

TIGER-Lab MMLU-Pro leaderboard (missing rows only) · Sep 22

83.5%

#24 / 45

Leader fable-5.1 · 92.4%
TIGER-Lab MMLU-Pro leaderboard (missing rows only) · Sep 22

GPQA Diamond

Artificial Analysis v4.3.2 · Oct 27

74.7%

#122 / 147

Leader gpt-6-astra · 96.1%
Artificial Analysis v4.3.2 · Oct 27

CritPt

Artificial Analysis v4.3.2 · Oct 27

0.3%

#134 / 159

Leader Muse Wonder Full · 32.9%
Artificial Analysis v4.3.2 · Oct 27

SciCode

Artificial Analysis model leaderboard · Sep 10

35.8%

#144 / 167

Leader opus-5.5 · 66.9%
Artificial Analysis model leaderboard · Sep 10

Humanity's Last Exam

Artificial Analysis v4.3.2 · Oct 27

7.0%

#152 / 167

Leader opus-5.5 · 61.4%
Artificial Analysis v4.3.2 · Oct 27

Not yet reported

MMMU-Pro

Programmatic Reasoning IQ 101 · measuredBest: #56 of 61 on LiveCodeBench1/6

BenchmarkScorePercentileRankLeader · source

LiveCodeBench

Vals.ai · Aug 4

50.3%

#56 / 61

Leader Fugu Ultra v1.0 · 93.2%
Vals.ai · Aug 4

Not yet reported

IOI V2Terminal-Bench 4.0ProgramBenchSWE-rebenchMirrorCode

Computer Use IQ 80 · measuredBest: #42 of 43 on BrowseComp1/5

BenchmarkScorePercentileRankLeader · source

BrowseComp

LLM Stats BrowseComp · Sep 22

9.9%

#42 / 43

Leader Kimi K3.1 Pro · 92.1%
LLM Stats BrowseComp · Sep 22

Not yet reported

OSWorld-VerifiedMCP AtlasArena.ai Agent ArenaAgents' Last Exam

Reliability IQ 110 · measuredBest: #45 of 96 on IFBench4/6

BenchmarkScorePercentileRankLeader · source

IFBench

Artificial Analysis v4.3.2 · Oct 27

70.3%

#45 / 96

Leader minimax-m3 · 82.9%
Artificial Analysis v4.3.2 · Oct 27

SimpleQA Verified

Full source leaderboard reconciliation · Oct 23

41.1%

#47 / 80

Leader Gemini 4 Argon · 76.0%
Full source leaderboard reconciliation · Oct 23

AA Omniscience

Artificial Analysis v4.3.2 · Oct 27

-11.0

#96 / 158

Leader opus-5.5 · 46.4
Artificial Analysis v4.3.2 · Oct 27

AA Long Context Reasoning v1.1

Artificial Analysis v4.3.2 · Oct 27

65.0%

#123 / 154

Leader kimi-k3 · 88.7%
Artificial Analysis v4.3.2 · Oct 27

Not yet reported

BullshitBench v2FACTS Grounding

Knowledge Work IQ 90 · estimatedNo published results0/6

No published results yet for Knowledge Work. The 90 shown is estimated from o1's other dimensions and will be replaced as results come in. How estimates work

AA-Briefcase v1.1GDPval-AA v2.1AutomationBench-AAGDP.pdfAA-AnalystAgentEnterpriseOps-Gym-AA

Release & availability

Release details on record for this model.

Released: Dec 5, 2024
Developer: OpenAI
Weights: ProprietaryHosted access only
API price: $15 in · $60 outper 1M tokens · full pricing

About this score

How much of o1's IQ is measured, and how the rest is estimated.

16/ 39benchmark results

6/ 7dimensions measured

1estimated

2/5Abstract

3/5Math

5/6Academic

1/6Coding

1/5Computer Use

4/6Reliability

0/6Knowledge Work

Bar height = share of benchmarks with resultsEstimatedNo score

Measured results are shown as published

Every score in the benchmark tables comes from its source unchanged. Composite IQ averages all seven dimensions equally.

Estimates fill the gaps

Knowledge Work has no published results yet, so its IQ is estimated from o1's other dimensions. Missing benchmarks can also draw on its predecessor, GPT 4o; a predecessor is an estimation input, not evidence that this model has the predecessor's measured results.

A missing result is not a zero

Benchmark versions and evaluation harnesses can differ between sources, so small gaps between models may not be meaningful.

Read the scoring methodology →

Compare o1

Side-by-side pages for the models it is most often weighed against.

#1 overall

gpt-6.1-sol

IQ141 46 higherEffective cost$6.30 11.8× cheaper

IQ95 same IQEffective cost$1.56 48.0× cheaper

Most similar profile

nova-2.0-pro

IQ97 2 higherEffective cost$6.90 10.8× cheaper

Similar price

fable-5

IQ133 38 higherEffective cost$87.20 1.2× pricier

#2 overall

opus-5.5

IQ141 46 higherEffective cost$35.73 2.1× cheaper

o1Comparison model

Compare with any modelBrowse all comparisons →

Frequently asked questions

Quick answers about o1, drawn from the same data as the rest of this page.

How AI IQ scores models →Get this data from the API →

What is the IQ of o1?: o1 has an estimated IQ of 95, which ranks #148 out of 176 public models on the AI IQ leaderboard. The score combines seven equally weighted reasoning dimensions using measured benchmark results and conservative estimates for missing coverage.
What is o1 best at?: Among the dimensions with published benchmark results, o1 ranks highest in Reliability (IQ 110, #108 of 177) and Programmatic Reasoning (IQ 101, #120 of 177). Its lowest-ranked measured dimension is Computer Use (IQ 80, #144 of 176).
What is the EQ of o1?: o1 doesn't have an EQ score yet. EQ (emotional reasoning) is calculated from EQ-Bench 4 and AttuneBench, and neither has published a result for o1. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.
Which benchmarks has o1 been tested on?: o1 has published results on 16 of the 39 benchmarks behind AI IQ's Composite IQ, including MathArena, IFBench, MMLU-Pro, SimpleQA Verified and AA Omniscience.
How much does o1 cost?: o1 costs $15.00 per million input tokens and $60.00 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $74.68 per million input-plus-output tokens.
Is o1 open source?: No — o1 is a proprietary model. Its weights are not publicly available, so it is used through hosted APIs and products.
Who developed o1?: o1 was developed by OpenAI, based in the United States.
When was o1 released?: o1 was publicly released on December 5, 2024.
How large is the context window of o1?: AI IQ hasn't verified an input context limit for o1 from its developer's documentation, so none is listed rather than guessing.
How fast is o1?: AI IQ doesn't have independent speed measurements for o1 yet. Speed comes from independent tests of a model's hosted API, which aren't available for every model — especially new, restricted-access, or self-hosted ones.