What is the IQ of Gemini 3.8 Flash?

Gemini 3.8 Flash has an estimated IQ of 130, which ranks #12 out of 169 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 3.8 Flash best at?

Among the dimensions with published benchmark results, Gemini 3.8 Flash ranks highest in Abstract Reasoning (IQ 126, #7 of 190) and Reliability (IQ 130, #7 of 169). Its lowest-ranked measured dimension is Mathematical Reasoning (IQ 128, #40 of 169).

What is the EQ of Gemini 3.8 Flash?

Gemini 3.8 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 3.8 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 3.8 Flash been tested on?

Gemini 3.8 Flash has published results on 27 of the 39 benchmarks behind AI IQ's Composite IQ, including Chess Puzzles, ARC-AGI-1, GPQA Diamond, MMMU-Pro and AA Omniscience.

How much does Gemini 3.8 Flash cost?

Gemini 3.8 Flash costs $0.75 per million input tokens and $3.75 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $7.99 per million input-plus-output tokens.

Is Gemini 3.8 Flash open source?

No — Gemini 3.8 Flash is a proprietary model. Its weights are not publicly available, so it is used through hosted APIs and products.

Who developed Gemini 3.8 Flash?

Gemini 3.8 Flash was developed by Google, based in the United States.

When was Gemini 3.8 Flash released?

Gemini 3.8 Flash was publicly released on September 2, 2026.

How large is the context window of Gemini 3.8 Flash?

Gemini 3.8 Flash supports a context window of 1,048,576 tokens — the amount of text it can consider in a single request.

How fast is Gemini 3.8 Flash?

In independent measurements, Gemini 3.8 Flash generates about 147 tokens per second, with a typical response time of about 26 seconds.

GoogleReleased Sep 2, 2026

Proprietary

Gemini 3.8 Flash

Strongest in Abstract Reasoning (#7 of 190) and Reliability (#7 of 169); weakest in Mathematical Reasoning (#40 of 169). All seven dimensions are measured.

Composite IQ ?

130

#12 of 169 ranked models · ahead of 93%

Google's #2 model, behind Gemini 4 Argon (140)

Effective cost

$7.99/1M

$0.75 in · $3.75 out

Speed

147tok/s

26s typical response

Context

1Mtokens

1,048,576 tokens in

Evidence

7of 7 dims

27 of 39 benchmarks

Compare with gpt-6.1-solAPI pricing

IQ profile

Seven dimensions, averaged equally into Composite IQ

Compare shape with

Measured95% 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

126measured

#7 / 190

5/5

-2.5

Mathematical ReasoningMath

128measured

#40 / 169

4/5

-18

Academic ReasoningAcademic

133measured

#15 / 192

5/6

-2.5

Programmatic ReasoningCoding

137measured

#20 / 170

3/6

-12

Computer UseComputer Use

129measured

#15 / 169

1/5

-11.5

ReliabilityReliability

130measured

#7 / 169

4/6

-2.5

Knowledge WorkKnowledge Work

127measured

#21 / 169

5/6

-4

Gemini 3.8 Flash95% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human

Benchmark results

27 of 39 Composite IQ benchmarks have a published result for Gemini 3.8 Flash. Benchmarks without a result stay listed so gaps are visible.

Abstract Reasoning IQ 126 · measuredBest: #2 of 76 on Chess Puzzles5/5

BenchmarkScorePercentileRankLeader · source

Chess Puzzles

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

61.0%

#2 / 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 · Oct 17

98.5%

#2 / 63

Leader fable-5 · 98.5%
ARC Prize verified leaderboard · Oct 17

ARC-AGI-2

ARC Prize verified leaderboard · Oct 17

89.2%

#10 / 63

Leader gpt-6-astra · 95.0%
ARC Prize verified leaderboard · Oct 17

Mystery Game Puzzles

Epoch AI vEpoch 100-position leaderboard (task implementation versions recorded per run) · Oct 17

47.0%

#11 / 55

Leader gpt-6-astra · 84.0%
Epoch AI vEpoch 100-position leaderboard (task implementation versions recorded per run) · Oct 17

ARC-AGI-3

ARC Prize verified leaderboard · Oct 17

10.4%

#5 / 22

Leader gpt-6-astra · 62.7%
ARC Prize verified leaderboard · Oct 17

Mathematical Reasoning IQ 128 · measuredBest: #25 of 82 on FrontierMath Tier 1-34/5

BenchmarkScorePercentileRankLeader · source

FrontierMath Tier 1-3

Epoch AI FrontierMath Tiers 1-3 v2 · Oct 17

68.4%

#25 / 82

Leader gpt-6-astra · 93.7%
Epoch AI FrontierMath Tiers 1-3 v2 · Oct 17

MathArena

MathArena Models · Oct 17

33.5%

#23 / 60

Leader gpt-6.1-sol · 84.0%
MathArena Models · Oct 17

FrontierMath Tier 4

Epoch AI FrontierMath Tier 4 v2 · Oct 17

22.0%

#40 / 70

Leader gpt-6.1-sol · 100%
Epoch AI FrontierMath Tier 4 v2 · Oct 17

ProofBench 1.1

ProofBench 1.1 · Oct 17

48.0%

#31 / 46

Leader fable-5.1 · 100%
ProofBench 1.1 · Oct 17

Not yet reported

AIME

Academic Reasoning IQ 133 · measuredBest: #7 of 146 on GPQA Diamond5/6

BenchmarkScorePercentileRankLeader · source

GPQA Diamond

Artificial Analysis v4.3.2 · Oct 17

95.3%

#7 / 146

Leader gpt-6-astra · 96.1%
Artificial Analysis v4.3.2 · Oct 17

MMMU-Pro

Artificial Analysis v4.3.2 · Oct 17

85.6%

#7 / 85

Leader fable-5.1 · 90.6%
Artificial Analysis v4.3.2 · Oct 17

Humanity's Last Exam

Artificial Analysis v4.3.2 · Oct 17

47.8%

#17 / 161

Leader opus-5.5 · 61.4%
Artificial Analysis v4.3.2 · Oct 17

SciCode

Artificial Analysis v4.3.2 · Oct 17

56.6%

#23 / 161

Leader opus-5.5 · 66.9%
Artificial Analysis v4.3.2 · Oct 17

CritPt

Artificial Analysis v4.3.2 · Oct 17

18.3%

#27 / 153

Leader Muse Wonder Full · 32.9%
Artificial Analysis v4.3.2 · Oct 17

Not yet reported

MMLU-Pro

Programmatic Reasoning IQ 137 · measuredBest: #27 of 99 on Terminal-Bench 4.03/6

BenchmarkScorePercentileRankLeader · source

Terminal-Bench 4.0

Artificial Analysis v4.3.2 · Oct 17

19.7%

#27 / 99

Leader sonnet-5.5 · 63.6%
Artificial Analysis v4.3.2 · Oct 17

ProgramBench

Vals.ai · Oct 17

14.0%

#22 / 61

Leader opus-5.5 · 65.0%
Vals.ai · Oct 17

IOI V2

Vals.ai v2 · Oct 17

56.9%

#18 / 41

Leader Gemini 4 Argon · 100%
Vals.ai v2 · Oct 17

Not yet reported

LiveCodeBenchSWE-rebenchMirrorCode

Computer Use IQ 129 · measuredBest: #14 of 59 on Arena.ai Agent Arena1/5

BenchmarkScorePercentileRankLeader · source

Arena.ai Agent Arena

Arena.ai Agent Arena · Oct 13

5.2

#14 / 59

Leader opus-5.5 · 17.9
Arena.ai Agent Arena · Oct 13

Not yet reported

BrowseCompOSWorld-VerifiedMCP AtlasAgents' Last Exam

Reliability IQ 130 · measuredBest: #13 of 152 on AA Omniscience4/6

BenchmarkScorePercentileRankLeader · source

AA Omniscience

Artificial Analysis v4.3.2 · Oct 17

29.6

#13 / 152

Leader opus-5.5 · 46.4
Artificial Analysis v4.3.2 · Oct 17

SimpleQA Verified

Epoch AI SimpleQA Verified · Oct 17

69.7%

#9 / 78

Leader Gemini 4 Argon · 76.0%
Epoch AI SimpleQA Verified · Oct 17

AA Long Context Reasoning v1.1

Artificial Analysis v4.3.2 · Oct 17

81.3%

#31 / 148

Leader kimi-k3 · 88.7%
Artificial Analysis v4.3.2 · Oct 17

FACTS Grounding

Kaggle FACTS Grounding v2 · Oct 17

73.1%

#13 / 36

Leader gpt-6-sol · 85.2%
Kaggle FACTS Grounding v2 · Oct 17

Not yet reported

BullshitBench v2IFBench

Knowledge Work IQ 127 · measuredBest: #16 of 101 on AutomationBench-AA5/6

BenchmarkScorePercentileRankLeader · source

AutomationBench-AA

Artificial Analysis v4.3.2 · Oct 17

59.9%

#16 / 101

Leader Gemini 4 Argon · 77.5%
Artificial Analysis v4.3.2 · Oct 17

GDP.pdf

Artificial Analysis v4.3.2 · Oct 17

21.0%

#21 / 100

Leader Muse Wonder Full · 32.6%
Artificial Analysis v4.3.2 · Oct 17

GDPval-AA v2.1 normalized score

Artificial Analysis v4.3.2 · Oct 17

46.8%

#34 / 119

Leader opus-5.5 · 68.3%
Artificial Analysis v4.3.2 · Oct 17

AA-Briefcase v1.1 Elo

Artificial Analysis v4.3.2 · Oct 17

1,203

#42 / 105

Leader sonnet-5.5 · 1,817
Artificial Analysis v4.3.2 · Oct 17

AA-AnalystAgent

Artificial Analysis vpass^5-2026-09-29 · Oct 17

46.3%

#14 / 30

Leader gemini-3.7-flash · 60.0%
Artificial Analysis vpass^5-2026-09-29 · Oct 17

Not yet reported

EnterpriseOps-Gym-AA

Release & availability

Facts from the publisher's announcement and documentation.

Released: Sep 2, 2026
Developer: Google
Access: Generally available through the Gemini API, Google AI Studio, Gemini Enterprise, and Google AI Pro and Ultra products
Weights: ProprietaryHosted access only
Inputs: textimagevideoaudiopdf
Context: 1,048,576 tokens
API price: $0.75 in · $3.75 outper 1M tokens · full pricing

What Google says

Google positions Gemini 3.8 Flash for long-horizon software engineering, autonomous agents, and complex enterprise workflows.
The model supports low, medium, and high reasoning effort, a 1,048,576-token input limit, and a 65,536-token output limit.
Artificial Analysis measured 304.6 output tokens per second at high effort.

Limitations

Google says the model can use more reasoning tokens than 3.7 Flash on complex tasks, especially at higher effort levels.
Computer use remains in preview, and the model does not support the Live API, image generation, or audio generation.

Publisher statements, not AI IQ measurements.

Sourcesblog.google/innovation-and-ai/models-and-research/gemini-models/3-8-flash-and-3-8-flash-cyber ai.google.dev/gemini-api/docs/models/gemini-3.8-flashartificialanalysis.ai/models/gemini-3-8-flash

About this score

How much of Gemini 3.8 Flash's IQ is measured, and how the rest is estimated.

27/ 39benchmark results

7/ 7dimensions measured

0estimated

5/5Abstract

4/5Math

5/6Academic

3/6Coding

1/5Computer Use

4/6Reliability

5/6Knowledge 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.

How gaps are handled

Every dimension has at least one published result. Missing benchmarks can also draw on its predecessors, Gemini 3.7 Flash and Gemini 3.6 Flash; 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 Gemini 3.8 Flash

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

#1 overall

gpt-6.1-sol

IQ141 11 higherEffective cost$6.30 1.3× cheaper

Predecessor

gemini-3.7-flash

IQ127 3 lowerEffective cost$5.94 1.3× cheaper

IQ130 same IQEffective cost$31.52 3.9× pricier

Most similar profile

DeepSeek V4.1 Flash

IQ126 4 lowerEffective cost$1.87 4.3× cheaper

Similar price

Qwen3.8 27B

IQ116 14 lowerEffective cost$7.74 about the same

Gemini 3.8 FlashComparison model

Compare with any modelBrowse all comparisons →

Frequently asked questions

Quick answers about Gemini 3.8 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 3.8 Flash?: Gemini 3.8 Flash has an estimated IQ of 130, which ranks #12 out of 169 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 3.8 Flash best at?: Among the dimensions with published benchmark results, Gemini 3.8 Flash ranks highest in Abstract Reasoning (IQ 126, #7 of 190) and Reliability (IQ 130, #7 of 169). Its lowest-ranked measured dimension is Mathematical Reasoning (IQ 128, #40 of 169).
What is the EQ of Gemini 3.8 Flash?: Gemini 3.8 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 3.8 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 3.8 Flash been tested on?: Gemini 3.8 Flash has published results on 27 of the 39 benchmarks behind AI IQ's Composite IQ, including Chess Puzzles, ARC-AGI-1, GPQA Diamond, MMMU-Pro and AA Omniscience.
How much does Gemini 3.8 Flash cost?: Gemini 3.8 Flash costs $0.75 per million input tokens and $3.75 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $7.99 per million input-plus-output tokens.
Is Gemini 3.8 Flash open source?: No — Gemini 3.8 Flash is a proprietary model. Its weights are not publicly available, so it is used through hosted APIs and products.
Who developed Gemini 3.8 Flash?: Gemini 3.8 Flash was developed by Google, based in the United States.
When was Gemini 3.8 Flash released?: Gemini 3.8 Flash was publicly released on September 2, 2026.
How large is the context window of Gemini 3.8 Flash?: Gemini 3.8 Flash supports a context window of 1,048,576 tokens — the amount of text it can consider in a single request.
How fast is Gemini 3.8 Flash?: In independent measurements, Gemini 3.8 Flash generates about 147 tokens per second, with a typical response time of about 26 seconds.