What is the IQ of llama-4-maverick?

llama-4-maverick has an estimated IQ of 87, which ranks #166 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 llama-4-maverick best at?

Among the dimensions with published benchmark results, llama-4-maverick ranks highest in Academic Reasoning (IQ 102, #161 of 203) and Knowledge Work (IQ 93, #145 of 176). Its lowest-ranked measured dimension is Abstract Reasoning (IQ 70, #201 of 201).

What is the EQ of llama-4-maverick?

llama-4-maverick has an estimated EQ of 90. EQ measures emotional reasoning — how well a model reads and responds to emotional context — and is scored separately from IQ.

Which benchmarks has llama-4-maverick been tested on?

llama-4-maverick has published results on 20 of the 39 benchmarks behind AI IQ's Composite IQ, including MMLU-Pro, BullshitBench v2, Terminal-Bench 4.0, IFBench and AA Omniscience.

How much does llama-4-maverick cost?

llama-4-maverick costs $0.26 per million input tokens and $0.91 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $0.0878 per million input-plus-output tokens.

Is llama-4-maverick open source?

Yes — llama-4-maverick is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.

Who developed llama-4-maverick?

llama-4-maverick was developed by Meta, based in the United States.

When was llama-4-maverick released?

llama-4-maverick was publicly released on April 5, 2025.

How large is the context window of llama-4-maverick?

llama-4-maverick supports a context window of 1,000,000 tokens — the amount of text it can consider in a single request.

How fast is llama-4-maverick?

In independent measurements, llama-4-maverick generates about 53 tokens per second, with a typical response time of about 10 seconds.

MetaReleased Apr 5, 2025

Open weights

llama-4-maverick

Its best results are in Academic Reasoning (#161 of 203) and Knowledge Work (#145 of 176); weakest in Abstract Reasoning (#201 of 201). Computer Use is estimated, because no results are published yet.

Composite IQ ?

#166 of 176 ranked models · ahead of 6%

Meta's #9 model, behind Muse Wonder Full (134)

Effective cost

$0.088/1M

$0.26 in · $0.91 out

Speed

53tok/s

10s typical response

Context

1Mtokens

1,000,000 tokens in

Evidence

6of 7 dims

20 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

70measured

#201 / 201

2/5

-56.5

Mathematical ReasoningMath

77measured

#173 / 176

2/5

-70

Academic ReasoningAcademic

102measured

#161 / 203

6/6

-33.5

Programmatic ReasoningCoding

97measured

#151 / 177

2/6

-52.5

Computer UseComputer Use

70estimated

#175 / 176

0/5

-73

ReliabilityReliability

100measured

#149 / 177

4/6

-34

Knowledge WorkKnowledge Work

93measured

#145 / 176

4/6

-38.5

llama-4-maverick95% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human

Benchmark results

20 of 39 Composite IQ benchmarks have a published result for llama-4-maverick. Benchmarks without a result stay listed so gaps are visible.

Abstract Reasoning IQ 70 · measuredBest: #67 of 68 on ARC-AGI-22/5

BenchmarkScorePercentileRankLeader · source

ARC-AGI-2

ARC Prize verified leaderboard · Sep 22

0.0%

#67 / 68

Leader gpt-6-astra · 95.0%
ARC Prize verified leaderboard · Sep 22

ARC-AGI-1

ARC Prize verified leaderboard · Sep 22

4.4%

#68 / 68

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

Not yet reported

ARC-AGI-3Chess PuzzlesMystery Game Puzzles

Mathematical Reasoning IQ 77 · measuredBest: #51 of 56 on AIME2/5

BenchmarkScorePercentileRankLeader · source

AIME

25.2%

#51 / 56

Leader gemini-3.1-pro · 98.1%

FrontierMath Tier 1-3

Epoch AI FrontierMath Tiers 1-3 v2 · Sep 22

0.7%

#83 / 84

Leader gpt-6-astra · 93.7%
Epoch AI FrontierMath Tiers 1-3 v2 · Sep 22

Not yet reported

FrontierMath Tier 4ProofBench 1.1MathArena

Academic Reasoning IQ 102 · measuredBest: #29 of 45 on MMLU-Pro6/6

BenchmarkScorePercentileRankLeader · source

MMLU-Pro

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

80.9%

#29 / 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

67.1%

#132 / 147

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

MMMU-Pro

Artificial Analysis v4.3.2 · Oct 27

62.1%

#80 / 89

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

SciCode

Artificial Analysis v4.3.2 · Oct 27

31.7%

#151 / 167

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

Humanity's Last Exam

Artificial Analysis v4.3.2 · Oct 27

4.9%

#157 / 167

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

CritPt

Artificial Analysis v4.3.2 · Oct 27

0.0%

#150 / 159

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

Programmatic Reasoning IQ 97 · measuredBest: #86 of 104 on Terminal-Bench 4.02/6

BenchmarkScorePercentileRankLeader · source

Terminal-Bench 4.0

Artificial Analysis v4.3.2 · Oct 27

0.0%

#86 / 104

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

LiveCodeBench

Vals.ai · Aug 4

47.3%

#57 / 61

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

Not yet reported

IOI V2ProgramBenchSWE-rebenchMirrorCode

Computer Use IQ 70 · estimatedNo published results0/5

No published results yet for Computer Use. The 70 shown is estimated from llama-4-maverick's other dimensions and will be replaced as results come in. How estimates work

BrowseCompOSWorld-VerifiedMCP AtlasArena.ai Agent ArenaAgents' Last Exam

Reliability IQ 100 · measuredBest: #54 of 83 on BullshitBench v24/6

BenchmarkScorePercentileRankLeader · source

BullshitBench v2

Full source leaderboard reconciliation · Oct 23

29.0%

#54 / 83

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

IFBench

Artificial Analysis v4.3.2 · Oct 27

43.0%

#85 / 96

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

AA Omniscience

Artificial Analysis v4.3.2 · Oct 27

-41.8

#141 / 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

50.0%

#138 / 154

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

Not yet reported

SimpleQA VerifiedFACTS Grounding

Knowledge Work IQ 93 · measuredBest: #113 of 124 on GDPval-AA v2.1 normalized score4/6

BenchmarkScorePercentileRankLeader · source

GDPval-AA v2.1 normalized score

Artificial Analysis v4.3.2 · Oct 23

0.0%

#113 / 124

Leader opus-5.5 · 68.3%
Artificial Analysis v4.3.2 · Oct 23

GDP.pdf

Artificial Analysis v4.3.2 · Oct 27

0.8%

#104 / 106

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

AutomationBench-AA

Artificial Analysis v4.3.2 · Oct 27

0.2%

#105 / 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

0.0

#110 / 111

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

Not yet reported

AA-AnalystAgentEnterpriseOps-Gym-AA

Release & availability

Release details on record for this model.

Released: Apr 5, 2025
Developer: Meta
Weights: Open weightsDownloadable and self-hostable
Context: 1,000,000 tokens
API price: $0.26 in · $0.91 outper 1M tokens · full pricing

About this score

How much of llama-4-maverick's IQ is measured, and how the rest is estimated.

20/ 39benchmark results

6/ 7dimensions measured

1estimated

2/5Abstract

2/5Math

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

Computer Use has no published results yet, so its IQ is estimated from llama-4-maverick'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 llama-4-maverick

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

#1 overall

gpt-6.1-sol

IQ141 54 higherEffective cost$6.30 71.8× pricier

IQ87 same IQEffective cost$0.208 2.4× pricier

Most similar profile

gemma-4-e4b

IQ88 1 higherEffective cost$0.026 3.4× cheaper

Similar price

qwen3.5-4b

IQ90 3 higherEffective cost$0.078 1.1× cheaper

#2 overall

opus-5.5

IQ141 54 higherEffective cost$35.73 406.9× pricier

llama-4-maverickComparison model

Compare with any modelBrowse all comparisons →

Frequently asked questions

Quick answers about llama-4-maverick, 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 llama-4-maverick?: llama-4-maverick has an estimated IQ of 87, which ranks #166 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 llama-4-maverick best at?: Among the dimensions with published benchmark results, llama-4-maverick ranks highest in Academic Reasoning (IQ 102, #161 of 203) and Knowledge Work (IQ 93, #145 of 176). Its lowest-ranked measured dimension is Abstract Reasoning (IQ 70, #201 of 201).
What is the EQ of llama-4-maverick?: llama-4-maverick has an estimated EQ of 90. EQ measures emotional reasoning — how well a model reads and responds to emotional context — and is scored separately from IQ.
Which benchmarks has llama-4-maverick been tested on?: llama-4-maverick has published results on 20 of the 39 benchmarks behind AI IQ's Composite IQ, including MMLU-Pro, BullshitBench v2, Terminal-Bench 4.0, IFBench and AA Omniscience.
How much does llama-4-maverick cost?: llama-4-maverick costs $0.26 per million input tokens and $0.91 per million output tokens. Adjusted for how many tokens it actually consumes on real tasks, its effective cost works out to $0.0878 per million input-plus-output tokens.
Is llama-4-maverick open source?: Yes — llama-4-maverick is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.
Who developed llama-4-maverick?: llama-4-maverick was developed by Meta, based in the United States.
When was llama-4-maverick released?: llama-4-maverick was publicly released on April 5, 2025.
How large is the context window of llama-4-maverick?: llama-4-maverick supports a context window of 1,000,000 tokens — the amount of text it can consider in a single request.
How fast is llama-4-maverick?: In independent measurements, llama-4-maverick generates about 53 tokens per second, with a typical response time of about 10 seconds.