What is the IQ of mistral-large-3?

mistral-large-3 has an estimated IQ of 91, which ranks #160 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 mistral-large-3 best at?

Among the dimensions with published benchmark results, mistral-large-3 ranks highest in Knowledge Work (IQ 96, #133 of 176) and Programmatic Reasoning (IQ 99, #141 of 177). Its lowest-ranked measured dimension is Reliability (IQ 90, #167 of 177).

What is the EQ of mistral-large-3?

mistral-large-3 has an estimated EQ of 101. EQ measures emotional reasoning — how well a model reads and responds to emotional context — and is scored separately from IQ.

Which benchmarks has mistral-large-3 been tested on?

mistral-large-3 has published results on 17 of the 39 benchmarks behind AI IQ's Composite IQ, including MMLU-Pro, SciCode, AIME, GPQA Diamond and AA Omniscience.

How much does mistral-large-3 cost?

mistral-large-3 costs $0.50 per million input tokens and $1.50 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $0.4404 per million input-plus-output tokens.

Is mistral-large-3 open source?

Yes — mistral-large-3 is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.

Who developed mistral-large-3?

mistral-large-3 was developed by Mistral, based in France.

When was mistral-large-3 released?

mistral-large-3 was publicly released on December 2, 2025.

How large is the context window of mistral-large-3?

mistral-large-3 supports a context window of 256,000 tokens — the amount of text it can consider in a single request.

How fast is mistral-large-3?

In independent measurements, mistral-large-3 generates about 86 tokens per second, with a typical response time of about 6.9 seconds.

MistralReleased Dec 2, 2025

Open weights

mistral-large-3

Its best results are in Knowledge Work (#133 of 176) and Programmatic Reasoning (#141 of 177); weakest in Reliability (#167 of 177). Abstract Reasoning and Computer Use are estimated, because no results are published yet.

Composite IQ ?

#160 of 176 ranked models · ahead of 9%

World (Other)'s #19 model, behind mistral-large-4 (119)

Effective cost

$0.44/1M

$0.50 in · $1.50 out

Speed

86tok/s

7s typical response

Context

256Ktokens

256,000 tokens in

Evidence

5of 7 dims

17 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

74estimated

#187 / 201

0/5

-52.5

Mathematical ReasoningMath

91measured

#154 / 176

1/5

-56

Academic ReasoningAcademic

101measured

#162 / 203

6/6

-34.5

Programmatic ReasoningCoding

99measured

#141 / 177

2/6

-50.5

Computer UseComputer Use

87estimated

#142 / 176

0/5

-56

ReliabilityReliability

90measured

#167 / 177

4/6

-44

Knowledge WorkKnowledge Work

96measured

#133 / 176

4/6

-35.5

mistral-large-395% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human

Benchmark results

17 of 39 Composite IQ benchmarks have a published result for mistral-large-3. Benchmarks without a result stay listed so gaps are visible.

Abstract Reasoning IQ 74 · estimatedNo published results0/5

No published results yet for Abstract Reasoning. The 74 shown is estimated from mistral-large-3's other dimensions and will be replaced as results come in. How estimates work

ARC-AGI-3ARC-AGI-2ARC-AGI-1Chess PuzzlesMystery Game Puzzles

Mathematical Reasoning IQ 91 · measuredBest: #48 of 56 on AIME1/5

BenchmarkScorePercentileRankLeader · source

AIME

42.9%

#48 / 56

Leader gemini-3.1-pro · 98.1%

Not yet reported

FrontierMath Tier 4FrontierMath Tier 1-3ProofBench 1.1MathArena

Academic Reasoning IQ 101 · measuredBest: #31 of 45 on MMLU-Pro6/6

BenchmarkScorePercentileRankLeader · source

MMLU-Pro

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

80.7%

#31 / 45

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

SciCode

Artificial Analysis v4.3.2 · Oct 27

36.6%

#140 / 167

Leader opus-5.5 · 66.9%
Artificial Analysis v4.3.2 · Oct 27

GPQA Diamond

Artificial Analysis v4.3.2 · Oct 27

68.0%

#129 / 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

4.2%

#159 / 167

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

MMMU-Pro

Artificial Analysis v4.3.2 · Oct 27

55.7%

#85 / 89

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

CritPt

Artificial Analysis v4.3.2 · Oct 27

0.0%

#152 / 159

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

Programmatic Reasoning IQ 99 · measuredBest: #55 of 61 on LiveCodeBench2/6

BenchmarkScorePercentileRankLeader · source

LiveCodeBench

Vals.ai · Aug 4

55.3%

#55 / 61

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

Terminal-Bench 4.0

Artificial Analysis v4.3.2 · Oct 27

0.0%

#94 / 104

Leader Kimi K3.1 Pro · 67.2%
Artificial Analysis v4.3.2 · Oct 27

Not yet reported

IOI V2ProgramBenchSWE-rebenchMirrorCode

Computer Use IQ 87 · estimatedNo published results0/5

No published results yet for Computer Use. The 87 shown is estimated from mistral-large-3's other dimensions and will be replaced as results come in. How estimates work

BrowseCompOSWorld-VerifiedMCP AtlasArena.ai Agent ArenaAgents' Last Exam

Reliability IQ 90 · measuredBest: #139 of 158 on AA Omniscience4/6

BenchmarkScorePercentileRankLeader · source

AA Omniscience

Artificial Analysis v4.3.2 · Oct 27

-39.6

#139 / 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

36.0%

#147 / 154

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

IFBench

Artificial Analysis v4.3.2 · Oct 27

36.2%

#92 / 96

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

BullshitBench v2

Full source leaderboard reconciliation · Oct 23

2.0%

#83 / 83

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

Not yet reported

SimpleQA VerifiedFACTS Grounding

Knowledge Work IQ 96 · measuredBest: #94 of 106 on AutomationBench-AA4/6

BenchmarkScorePercentileRankLeader · source

AutomationBench-AA

Artificial Analysis v4.3.2 · Oct 27

1.5%

#94 / 106

Leader Gemini 4 Argon · 77.5%
Artificial Analysis v4.3.2 · Oct 27

AA-Briefcase v1.1 Elo

Artificial Analysis v4.3.2 · Oct 27

228.1

#102 / 111

Leader sonnet-5.5 · 1,812
Artificial Analysis v4.3.2 · Oct 27

GDP.pdf

Artificial Analysis v4.3.2 · Oct 27

1.2%

#99 / 106

Leader Muse Wonder Full · 32.6%
Artificial Analysis v4.3.2 · Oct 27

GDPval-AA v2.1 normalized score

Artificial Analysis v4.3.2 · Oct 20

0.0%

#117 / 124

Leader opus-5.5 · 68.3%
Artificial Analysis v4.3.2 · Oct 20

Not yet reported

AA-AnalystAgentEnterpriseOps-Gym-AA

Release & availability

Release details on record for this model.

Released: Dec 2, 2025
Developer: Mistral
Weights: Open weightsDownloadable and self-hostable
Context: 256,000 tokens
API price: $0.50 in · $1.50 outper 1M tokens · full pricing

About this score

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

17/ 39benchmark results

5/ 7dimensions measured

2estimated

0/5Abstract

1/5Math

6/6Academic

2/6Coding

0/5Computer Use

4/6Reliability

4/6Knowledge 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

Abstract Reasoning and Computer Use have no published results yet, so their IQ is estimated from mistral-large-3'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 mistral-large-3

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

#1 overall

gpt-6.1-sol

IQ141 50 higherEffective cost$6.30 14.3× pricier

World (Other) #1

mistral-large-4

IQ119 28 higherEffective cost$4.69 10.7× pricier

Closest IQ

kimi-k2

IQ91 same IQEffective cost$2.38 5.4× pricier

Most similar profile

Granite 4.2 8B

IQ94 3 higherEffective cost$0.25 1.8× cheaper

Similar price

gemini-3.1-flash-lite

IQ104 13 higherEffective cost$0.429 about the same

#2 overall

opus-5.5

IQ141 50 higherEffective cost$35.73 81.1× pricier

mistral-large-3Comparison model

Compare with any modelBrowse all comparisons →

Frequently asked questions

Quick answers about mistral-large-3, 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 mistral-large-3?: mistral-large-3 has an estimated IQ of 91, which ranks #160 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 mistral-large-3 best at?: Among the dimensions with published benchmark results, mistral-large-3 ranks highest in Knowledge Work (IQ 96, #133 of 176) and Programmatic Reasoning (IQ 99, #141 of 177). Its lowest-ranked measured dimension is Reliability (IQ 90, #167 of 177).
What is the EQ of mistral-large-3?: mistral-large-3 has an estimated EQ of 101. EQ measures emotional reasoning — how well a model reads and responds to emotional context — and is scored separately from IQ.
Which benchmarks has mistral-large-3 been tested on?: mistral-large-3 has published results on 17 of the 39 benchmarks behind AI IQ's Composite IQ, including MMLU-Pro, SciCode, AIME, GPQA Diamond and AA Omniscience.
How much does mistral-large-3 cost?: mistral-large-3 costs $0.50 per million input tokens and $1.50 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $0.4404 per million input-plus-output tokens.
Is mistral-large-3 open source?: Yes — mistral-large-3 is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.
Who developed mistral-large-3?: mistral-large-3 was developed by Mistral, based in France.
When was mistral-large-3 released?: mistral-large-3 was publicly released on December 2, 2025.
How large is the context window of mistral-large-3?: mistral-large-3 supports a context window of 256,000 tokens — the amount of text it can consider in a single request.
How fast is mistral-large-3?: In independent measurements, mistral-large-3 generates about 86 tokens per second, with a typical response time of about 6.9 seconds.