What is the IQ of gpt-oss-20b?

gpt-oss-20b has an estimated IQ of 93, which ranks #155 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 gpt-oss-20b best at?

Among the dimensions with published benchmark results, gpt-oss-20b ranks highest in Programmatic Reasoning (IQ 115, #76 of 177) and Mathematical Reasoning (IQ 107, #102 of 176). Its lowest-ranked measured dimension is Academic Reasoning (IQ 99, #194 of 203).

What is the EQ of gpt-oss-20b?

gpt-oss-20b has an estimated EQ of 93. EQ measures emotional reasoning — how well a model reads and responds to emotional context — and is scored separately from IQ.

Which benchmarks has gpt-oss-20b been tested on?

gpt-oss-20b has published results on 13 of the 39 benchmarks behind AI IQ's Composite IQ, including AIME, LiveCodeBench, IFBench, SciCode and MMLU-Pro.

How much does gpt-oss-20b cost?

gpt-oss-20b costs $0.07 per million input tokens and $0.215 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $0.2447 per million input-plus-output tokens.

Is gpt-oss-20b open source?

Yes — gpt-oss-20b is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.

Who developed gpt-oss-20b?

gpt-oss-20b was developed by OpenAI, based in the United States.

When was gpt-oss-20b released?

gpt-oss-20b was publicly released on August 5, 2025.

How large is the context window of gpt-oss-20b?

gpt-oss-20b supports a context window of 131,000 tokens — the amount of text it can consider in a single request.

How fast is gpt-oss-20b?

In independent measurements, gpt-oss-20b generates about 243 tokens per second, with a typical response time of about 11 seconds.

OpenAIReleased Aug 5, 2025

Open weights

gpt-oss-20b

Its best results are in Programmatic Reasoning (#76 of 177) and Mathematical Reasoning (#102 of 176); weakest in Academic Reasoning (#194 of 203). Computer Use and Knowledge Work are estimated, because no results are published yet.

Composite IQ ?

#155 of 176 ranked models · ahead of 12%

OpenAI's #21 model, behind gpt-6.1-sol (141)

Effective cost

$0.245/1M

$0.07 in · $0.21 out

Speed

243tok/s

11s typical response

Context

131Ktokens

131,000 tokens in

Evidence

5of 7 dims

13 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

74measured

#186 / 201

1/5

-52.5

Mathematical ReasoningMath

107measured

#102 / 176

2/5

-40

Academic ReasoningAcademic

99measured

#194 / 203

5/6

-36.5

Programmatic ReasoningCoding

115measured

#76 / 177

1/6

-34.5

Computer UseComputer Use

70estimated

#171 / 176

0/5

-73

ReliabilityReliability

91measured

#165 / 177

4/6

-43

Knowledge WorkKnowledge Work

92estimated

#147 / 176

0/6

-39.5

gpt-oss-20b95% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human

Benchmark results

13 of 39 Composite IQ benchmarks have a published result for gpt-oss-20b. Benchmarks without a result stay listed so gaps are visible.

Abstract Reasoning IQ 74 · measuredBest: #75 of 76 on Chess Puzzles1/5

BenchmarkScorePercentileRankLeader · source

Chess Puzzles

Epoch AI vEpoch 100-position leaderboard (task implementation versions recorded per run) · Oct 15

4.0%

#75 / 76

Leader gpt-6-astra · 72.0%
Epoch AI vEpoch 100-position leaderboard (task implementation versions recorded per run) · Oct 15

Not yet reported

ARC-AGI-3ARC-AGI-2ARC-AGI-1Mystery Game Puzzles

Mathematical Reasoning IQ 107 · measuredBest: #31 of 56 on AIME2/5

BenchmarkScorePercentileRankLeader · source

AIME

86.0%

#31 / 56

Leader gemini-3.1-pro · 98.1%

MathArena

Full source leaderboard reconciliation · Oct 28

12.4%

#57 / 64

Leader gpt-6.1-sol · 84.0%
Full source leaderboard reconciliation · Oct 28

Not yet reported

FrontierMath Tier 4FrontierMath Tier 1-3ProofBench 1.1

Academic Reasoning IQ 99 · measuredBest: #147 of 167 on SciCode5/6

BenchmarkScorePercentileRankLeader · source

SciCode

Artificial Analysis model leaderboard · Sep 10

34.4%

#147 / 167

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

MMLU-Pro

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

74.8%

#40 / 45

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

CritPt

Artificial Analysis v4.3.2 · Oct 27

0.0%

#143 / 159

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

GPQA Diamond

Artificial Analysis v4.3.2 · Oct 27

61.1%

#136 / 147

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

Humanity's Last Exam

Artificial Analysis v4.3.2 · Oct 27

5.3%

#156 / 167

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

Not yet reported

MMMU-Pro

Programmatic Reasoning IQ 115 · measuredBest: #37 of 61 on LiveCodeBench1/6

BenchmarkScorePercentileRankLeader · source

LiveCodeBench

Vals.ai · Aug 4

80.4%

#37 / 61

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

Not yet reported

IOI V2Terminal-Bench 4.0ProgramBenchSWE-rebenchMirrorCode

Computer Use IQ 70 · estimatedNo published results0/5

No published results yet for Computer Use. The 70 shown is estimated from gpt-oss-20b's other dimensions and will be replaced as results come in. How estimates work

BrowseCompOSWorld-VerifiedMCP AtlasArena.ai Agent ArenaAgents' Last Exam

Reliability IQ 91 · measuredBest: #71 of 96 on IFBench4/6

BenchmarkScorePercentileRankLeader · source

IFBench

Artificial Analysis v4.3.2 · Oct 27

57.8%

#71 / 96

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

AA Long Context Reasoning v1.1

Artificial Analysis v4.3.2 · Oct 27

31.0%

#150 / 154

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

AA Omniscience

Artificial Analysis v4.3.2 · Oct 27

-58.5

#157 / 158

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

FACTS Grounding

Full source leaderboard reconciliation · Oct 23

23.1%

#36 / 36

Leader gpt-6-sol · 85.2%
Full source leaderboard reconciliation · Oct 23

Not yet reported

SimpleQA VerifiedBullshitBench v2

Knowledge Work IQ 92 · estimatedNo published results0/6

No published results yet for Knowledge Work. The 92 shown is estimated from gpt-oss-20b'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: Aug 5, 2025
Developer: OpenAI
Weights: Open weightsDownloadable and self-hostable
Context: 131,000 tokens
API price: $0.07 in · $0.21 outper 1M tokens · full pricing

About this score

How much of gpt-oss-20b's IQ is measured, and how the rest is estimated.

13/ 39benchmark results

5/ 7dimensions measured

2estimated

1/5Abstract

2/5Math

5/6Academic

1/6Coding

0/5Computer 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

Computer Use and Knowledge Work have no published results yet, so their IQ is estimated from gpt-oss-20b's other dimensions.

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 gpt-oss-20b

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

#1 overall

gpt-6.1-sol

IQ141 48 higherEffective cost$6.30 25.8× pricier

Closest IQ

trinity-large

IQ93 same IQEffective cost$1.04 4.3× pricier

Most similar profile

intellect-3

IQ95 2 higherEffective cost$1.56 6.4× pricier

Similar price

mimo-v2.5

IQ108 15 higherEffective cost$0.243 about the same

#2 overall

opus-5.5

IQ141 48 higherEffective cost$35.73 146.0× pricier

#3 overall

Gemini 4 Argon

IQ140 47 higher

gpt-oss-20bComparison model

Compare with any modelBrowse all comparisons →

Frequently asked questions

Quick answers about gpt-oss-20b, 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 gpt-oss-20b?: gpt-oss-20b has an estimated IQ of 93, which ranks #155 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 gpt-oss-20b best at?: Among the dimensions with published benchmark results, gpt-oss-20b ranks highest in Programmatic Reasoning (IQ 115, #76 of 177) and Mathematical Reasoning (IQ 107, #102 of 176). Its lowest-ranked measured dimension is Academic Reasoning (IQ 99, #194 of 203).
What is the EQ of gpt-oss-20b?: gpt-oss-20b has an estimated EQ of 93. EQ measures emotional reasoning — how well a model reads and responds to emotional context — and is scored separately from IQ.
Which benchmarks has gpt-oss-20b been tested on?: gpt-oss-20b has published results on 13 of the 39 benchmarks behind AI IQ's Composite IQ, including AIME, LiveCodeBench, IFBench, SciCode and MMLU-Pro.
How much does gpt-oss-20b cost?: gpt-oss-20b costs $0.07 per million input tokens and $0.215 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $0.2447 per million input-plus-output tokens.
Is gpt-oss-20b open source?: Yes — gpt-oss-20b is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.
Who developed gpt-oss-20b?: gpt-oss-20b was developed by OpenAI, based in the United States.
When was gpt-oss-20b released?: gpt-oss-20b was publicly released on August 5, 2025.
How large is the context window of gpt-oss-20b?: gpt-oss-20b supports a context window of 131,000 tokens — the amount of text it can consider in a single request.
How fast is gpt-oss-20b?: In independent measurements, gpt-oss-20b generates about 243 tokens per second, with a typical response time of about 11 seconds.