What is the IQ of gemini-3-flash?

gemini-3-flash has an estimated IQ of 111, which ranks #72 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 gemini-3-flash best at?

Among the dimensions with published benchmark results, gemini-3-flash ranks highest in Academic Reasoning (IQ 127, #61 of 203) and Abstract Reasoning (IQ 99, #63 of 201). Its lowest-ranked measured dimension is Programmatic Reasoning (IQ 103, #114 of 177).

What is the EQ of gemini-3-flash?

gemini-3-flash 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 gemini-3-flash. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.

Which benchmarks has gemini-3-flash been tested on?

gemini-3-flash has published results on 24 of the 39 benchmarks behind AI IQ's Composite IQ, including IFBench, SimpleQA Verified, MMLU-Pro, AIME and LiveCodeBench.

How much does gemini-3-flash cost?

gemini-3-flash costs $0.50 per million input tokens and $3.00 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $2.60 per million input-plus-output tokens.

Is gemini-3-flash open source?

No — gemini-3-flash is a proprietary model. Its weights are not publicly available, so it is used through hosted APIs and products.

Who developed gemini-3-flash?

gemini-3-flash was developed by Google, based in the United States.

When was gemini-3-flash released?

gemini-3-flash was publicly released on December 17, 2025.

How large is the context window of gemini-3-flash?

AI IQ hasn't verified an input context limit for gemini-3-flash from its developer's documentation, so none is listed rather than guessing.

How fast is gemini-3-flash?

In independent measurements, gemini-3-flash generates about 216 tokens per second, with a typical response time of about 8.1 seconds.

GoogleReleased Dec 17, 2025

Proprietary

gemini-3-flash

Strongest in Academic Reasoning (#61 of 203) and Abstract Reasoning (#63 of 201); weakest in Programmatic Reasoning (#114 of 177). Knowledge Work is estimated, because no results are published yet.

Composite IQ ?

111

#72 of 176 ranked models · ahead of 59%

Google's #9 model, behind Gemini 4 Argon (140)

Effective cost

$2.60/1M

$0.50 in · $3 out

Speed

216tok/s

8s typical response

Context

Unverified

Not in publisher docs

Evidence

6of 7 dims

24 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

99measured

#63 / 201

4/5

-27.5

Mathematical ReasoningMath

118measured

#66 / 176

4/5

-29

Academic ReasoningAcademic

127measured

#61 / 203

6/6

-8.5

Programmatic ReasoningCoding

103measured

#114 / 177

2/6

-46.5

Computer UseComputer Use

113measured

#79 / 176

2/5

-30

ReliabilityReliability

116measured

#63 / 177

6/6

-18

Knowledge WorkKnowledge Work

97estimated

#126 / 176

0/6

-34.5

gemini-3-flash95% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human

Benchmark results

24 of 39 Composite IQ benchmarks have a published result for gemini-3-flash. Benchmarks without a result stay listed so gaps are visible.

Abstract Reasoning IQ 99 · measuredBest: #17 of 76 on Chess Puzzles4/5

BenchmarkScorePercentileRankLeader · source

Chess Puzzles

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

40.0%

#17 / 76

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

Mystery Game Puzzles

Full source leaderboard reconciliation vEpoch 100-position leaderboard (task implementation versions recorded per run) · Oct 23

26.0%

#31 / 55

Leader gpt-6-astra · 84.0%
Full source leaderboard reconciliation vEpoch 100-position leaderboard (task implementation versions recorded per run) · Oct 23

ARC-AGI-1

Full source leaderboard reconciliation · Oct 23

84.7%

#39 / 68

Leader fable-5 · 98.5%
Full source leaderboard reconciliation · Oct 23

ARC-AGI-2

Full source leaderboard reconciliation · Oct 23

33.6%

#42 / 68

Leader gpt-6-astra · 95.0%
Full source leaderboard reconciliation · Oct 23

Not yet reported

ARC-AGI-3

Mathematical Reasoning IQ 118 · measuredBest: #9 of 56 on AIME4/5

BenchmarkScorePercentileRankLeader · source

AIME

95.6%

#9 / 56

Leader gemini-3.1-pro · 98.1%

FrontierMath Tier 1-3

Full source leaderboard reconciliation · Oct 23

51.2%

#48 / 84

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

MathArena

Full source leaderboard reconciliation · Oct 28

20.0%

#40 / 64

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

FrontierMath Tier 4

Full source leaderboard reconciliation · Oct 23

17.1%

#46 / 72

Leader gpt-6.1-sol · 100%
Full source leaderboard reconciliation · Oct 23

Not yet reported

ProofBench 1.1

Academic Reasoning IQ 127 · measuredBest: #7 of 45 on MMLU-Pro6/6

BenchmarkScorePercentileRankLeader · source

MMLU-Pro

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

89.0%

#7 / 45

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

MMMU-Pro

Artificial Analysis v4.3.2 · Oct 27

79.9%

#27 / 89

Leader fable-5.1 · 90.6%
Artificial Analysis v4.3.2 · Oct 27

GPQA Diamond

Artificial Analysis v4.3.2 · Oct 27

89.8%

#48 / 147

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

SciCode

Artificial Analysis model leaderboard · Sep 10

50.6%

#60 / 167

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

Humanity's Last Exam

Artificial Analysis v4.3.2 · Oct 27

36.6%

#64 / 167

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

CritPt

Artificial Analysis v4.3.2 · Oct 27

8.6%

#69 / 159

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

Programmatic Reasoning IQ 103 · measuredBest: #13 of 61 on LiveCodeBench2/6

BenchmarkScorePercentileRankLeader · source

LiveCodeBench

Vals.ai · Aug 4

85.6%

#13 / 61

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

ProgramBench

Full source leaderboard reconciliation · Oct 23

0.0%

#52 / 61

Leader opus-5.5 · 65.0%
Full source leaderboard reconciliation · Oct 23

Not yet reported

IOI V2Terminal-Bench 4.0SWE-rebenchMirrorCode

Computer Use IQ 113 · measuredBest: #27 of 32 on MCP Atlas2/5

BenchmarkScorePercentileRankLeader · source

MCP Atlas

Full source leaderboard reconciliation · Oct 23

62.0%

#27 / 32

Leader muse-spark-1.1 · 88.1%
Full source leaderboard reconciliation · Oct 23

Arena.ai Agent Arena

Arena.ai Agent Arena · Sep 22

-8.3

#60 / 70

Leader opus-5.5 · 20.9
Arena.ai Agent Arena · Sep 22

Not yet reported

BrowseCompOSWorld-VerifiedAgents' Last Exam

Reliability IQ 116 · measuredBest: #10 of 96 on IFBench6/6

BenchmarkScorePercentileRankLeader · source

IFBench

Artificial Analysis v4.3.2 · Oct 27

78.0%

#10 / 96

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

SimpleQA Verified

Full source leaderboard reconciliation · Oct 23

66.8%

#12 / 80

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

AA Omniscience

Artificial Analysis v4.3.2 · Oct 27

10.1

#45 / 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

78.0%

#69 / 154

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

FACTS Grounding

Full source leaderboard reconciliation · Oct 23

59.0%

#27 / 36

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

BullshitBench v2

Full source leaderboard reconciliation · Oct 23

10.0%

#73 / 83

Leader opus-4.8 · 94.0%
Full source leaderboard reconciliation · Oct 23

Knowledge Work IQ 97 · estimatedNo published results0/6

No published results yet for Knowledge Work. The 97 shown is estimated from gemini-3-flash'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 17, 2025
Developer: Google
Weights: ProprietaryHosted access only
API price: $0.50 in · $3 outper 1M tokens · full pricing

About this score

How much of gemini-3-flash's IQ is measured, and how the rest is estimated.

24/ 39benchmark results

6/ 7dimensions measured

1estimated

4/5Abstract

4/5Math

6/6Academic

2/6Coding

2/5Computer Use

6/6Reliability

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 gemini-3-flash'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 gemini-3-flash

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

#1 overall

gpt-6.1-sol

IQ141 30 higherEffective cost$6.30 2.4× pricier

IQ111 same IQEffective cost$3.57 1.4× pricier

Most similar profile

grok-4.20

IQ111 same IQEffective cost$5.12 2.0× pricier

Similar price

deepseek-v3.1

IQ98 13 lowerEffective cost$2.48 about the same

#2 overall

opus-5.5

IQ141 30 higherEffective cost$35.73 13.8× pricier

gemini-3-flashComparison model

Compare with any modelBrowse all comparisons →

Frequently asked questions

Quick answers about gemini-3-flash, 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 gemini-3-flash?: gemini-3-flash has an estimated IQ of 111, which ranks #72 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 gemini-3-flash best at?: Among the dimensions with published benchmark results, gemini-3-flash ranks highest in Academic Reasoning (IQ 127, #61 of 203) and Abstract Reasoning (IQ 99, #63 of 201). Its lowest-ranked measured dimension is Programmatic Reasoning (IQ 103, #114 of 177).
What is the EQ of gemini-3-flash?: gemini-3-flash 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 gemini-3-flash. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.
Which benchmarks has gemini-3-flash been tested on?: gemini-3-flash has published results on 24 of the 39 benchmarks behind AI IQ's Composite IQ, including IFBench, SimpleQA Verified, MMLU-Pro, AIME and LiveCodeBench.
How much does gemini-3-flash cost?: gemini-3-flash costs $0.50 per million input tokens and $3.00 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $2.60 per million input-plus-output tokens.
Is gemini-3-flash open source?: No — gemini-3-flash is a proprietary model. Its weights are not publicly available, so it is used through hosted APIs and products.
Who developed gemini-3-flash?: gemini-3-flash was developed by Google, based in the United States.
When was gemini-3-flash released?: gemini-3-flash was publicly released on December 17, 2025.
How large is the context window of gemini-3-flash?: AI IQ hasn't verified an input context limit for gemini-3-flash from its developer's documentation, so none is listed rather than guessing.
How fast is gemini-3-flash?: In independent measurements, gemini-3-flash generates about 216 tokens per second, with a typical response time of about 8.1 seconds.