What is the IQ of Muse Spark Open?

Muse Spark Open has an estimated IQ of 122, which ranks #37 out of 168 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 Muse Spark Open best at?

Among the dimensions with published benchmark results, Muse Spark Open ranks highest in Reliability (IQ 126, #20 of 168) and Academic Reasoning (IQ 130, #30 of 187). Its lowest-ranked measured dimension is Programmatic Reasoning (IQ 126, #42 of 169).

What is the EQ of Muse Spark Open?

Muse Spark Open 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 Muse Spark Open. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.

Which benchmarks has Muse Spark Open been tested on?

Muse Spark Open has published results on 15 of the 39 benchmarks behind AI IQ's Composite IQ, including SciCode, AA Omniscience, Humanity's Last Exam, MMMU-Pro and CritPt.

How much does Muse Spark Open cost?

AI IQ doesn't have a verified standard per-token API price for Muse Spark Open, so it can't calculate an effective cost. This is common for models offered only through subscriptions, restricted-access programs, or self-hosting.

Is Muse Spark Open open source?

Yes — Muse Spark Open is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.

Who developed Muse Spark Open?

Muse Spark Open was developed by Meta, based in the United States.

When was Muse Spark Open released?

Muse Spark Open was publicly released on October 15, 2026.

How large is the context window of Muse Spark Open?

Muse Spark Open supports a context window of 1,000,000 tokens — the amount of text it can consider in a single request.

How fast is Muse Spark Open?

In independent measurements, Muse Spark Open generates about 103 tokens per second, with a typical response time of about 25 seconds.

MetaReleased Oct 15, 2026

NewOpen weights

Muse Spark Open

Strongest in Reliability (#20 of 168) and Academic Reasoning (#30 of 187); weakest in Programmatic Reasoning (#42 of 169). Abstract Reasoning and Computer Use are estimated, because no results are published yet.

Composite IQ ?

122

#37 of 168 ranked models · ahead of 78%

Meta's #4 model, behind Muse Wonder Full (131)

Effective cost

Not priced

No standard API rate

Speed

103tok/s

25s typical response

Context

1Mtokens

1,000,000 tokens in

Evidence

5of 7 dims

15 of 39 benchmarks

Compare with gpt-6.1-sol

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

102estimated

#49 / 185

0/5

-26.5

Mathematical ReasoningMath

130measured

#29 / 168

2/5

-16

Academic ReasoningAcademic

130measured

#30 / 187

5/6

-5.5

Programmatic ReasoningCoding

126measured

#42 / 169

1/6

-23

Computer UseComputer Use

118estimated

#55 / 168

0/5

-22.5

ReliabilityReliability

126measured

#20 / 168

3/6

-6.5

Knowledge WorkKnowledge Work

124measured

#33 / 168

4/6

-7

Muse Spark Open95% range (model-based, uncalibrated)Top-10 median#1 in dimension100 = average human

Benchmark results

15 of 39 Composite IQ benchmarks have a published result for Muse Spark Open. Benchmarks without a result stay listed so gaps are visible.

Abstract Reasoning IQ 102 · estimatedNo published results0/5

No published results yet for Abstract Reasoning. The 102 shown is estimated from Muse Spark Open'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 130 · measuredBest: #23 of 82 on FrontierMath Tier 1-32/5

BenchmarkScorePercentileRankLeader · source

FrontierMath Tier 1-3

Epoch AI FrontierMath Tiers 1-3 v2 · Oct 16

69.8%

#23 / 82

Leader gpt-6-astra · 93.7%
Epoch AI FrontierMath Tiers 1-3 v2 · Oct 16

FrontierMath Tier 4

Epoch AI FrontierMath Tier 4 v2 · Oct 16

31.7%

#25 / 70

Leader gpt-6.1-sol · 100%
Epoch AI FrontierMath Tier 4 v2 · Oct 16

Not yet reported

ProofBench 1.1MathArenaAIME

Academic Reasoning IQ 130 · measuredBest: #17 of 161 on SciCode5/6

BenchmarkScorePercentileRankLeader · source

SciCode

Artificial Analysis v4.3.2 · Oct 16

58.4%

#17 / 161

Leader opus-5.5 · 66.9%
Artificial Analysis v4.3.2 · Oct 16

Humanity's Last Exam

Artificial Analysis v4.3.2 · Oct 16

43.9%

#25 / 161

Leader opus-5.5 · 61.4%
Artificial Analysis v4.3.2 · Oct 16

MMMU-Pro

Artificial Analysis v4.3.2 · Oct 16

80.9%

#17 / 85

Leader fable-5.1 · 90.6%
Artificial Analysis v4.3.2 · Oct 16

CritPt

Artificial Analysis v4.3.2 · Oct 16

14.9%

#41 / 153

Leader Muse Wonder Full · 32.9%
Artificial Analysis v4.3.2 · Oct 16

GPQA Diamond

Artificial Analysis v4.3.2 · Oct 16

89.0%

#57 / 146

Leader gpt-6-astra · 96.1%
Artificial Analysis v4.3.2 · Oct 16

Not yet reported

MMLU-Pro

Programmatic Reasoning IQ 126 · measuredBest: #39 of 99 on Terminal-Bench 4.01/6

BenchmarkScorePercentileRankLeader · source

Terminal-Bench 4.0

Artificial Analysis v4.3.2 · Oct 16

8.1%

#39 / 99

Leader sonnet-5.5 · 63.6%
Artificial Analysis v4.3.2 · Oct 16

Not yet reported

LiveCodeBenchIOI V2ProgramBenchSWE-rebenchMirrorCode

Computer Use IQ 118 · estimatedNo published results0/5

No published results yet for Computer Use. The 118 shown is estimated from Muse Spark Open's other dimensions and will be replaced as results come in. How estimates work

BrowseCompOSWorld-VerifiedMCP AtlasArena.ai Agent ArenaAgents' Last Exam

Reliability IQ 126 · measuredBest: #22 of 152 on AA Omniscience3/6

BenchmarkScorePercentileRankLeader · source

AA Omniscience

Artificial Analysis v4.3.2 · Oct 16

24.7

#22 / 152

Leader opus-5.5 · 46.4
Artificial Analysis v4.3.2 · Oct 16

SimpleQA Verified

Epoch AI SimpleQA Verified · Oct 16

56.6%

#21 / 78

Leader Gemini 4 Argon · 76.0%
Epoch AI SimpleQA Verified · Oct 16

AA Long Context Reasoning v1.1

Artificial Analysis v4.3.2 · Oct 16

78.3%

#59 / 148

Leader kimi-k3 · 88.7%
Artificial Analysis v4.3.2 · Oct 16

Not yet reported

BullshitBench v2IFBenchFACTS Grounding

Knowledge Work IQ 124 · measuredBest: #28 of 100 on GDP.pdf4/6

BenchmarkScorePercentileRankLeader · source

GDP.pdf

Artificial Analysis v4.3.2 · Oct 16

18.6%

#28 / 100

Leader Muse Wonder Full · 32.6%
Artificial Analysis v4.3.2 · Oct 16

GDPval-AA v2.1 normalized score

Artificial Analysis v4.3.2 · Oct 16

45.1%

#39 / 119

Leader opus-5.5 · 68.3%
Artificial Analysis v4.3.2 · Oct 16

AutomationBench-AA

Artificial Analysis v4.3.2 · Oct 16

49.7%

#34 / 101

Leader Gemini 4 Argon · 77.5%
Artificial Analysis v4.3.2 · Oct 16

AA-Briefcase v1.1 Elo

Artificial Analysis v4.3.2 · Oct 16

1,279

#37 / 105

Leader sonnet-5.5 · 1,817
Artificial Analysis v4.3.2 · Oct 16

Not yet reported

AA-AnalystAgentEnterpriseOps-Gym-AA

Release & availability

Facts from the publisher's announcement and documentation.

Released: Oct 15, 2026
Developer: Meta
Access: Downloadable weights under the custom Muse Model License, with hosted access through inference partners.
Weights: Open weightsDownloadable and self-hostable
Inputs: textimagevideoaudio
Context: 1,000,000 tokens
API price: No per-token rateNo current standard hosted token price has been verified for this exact checkpoint; weight availability is separate from hosted API pricing.

What Meta says

Meta announced this distinct open-weight model on October 15, 2026.
Mixture-of-experts architecture with 816B total and 42B active parameters; designed for coding and agentic work.

Limitations

Custom Muse Model License includes attribution, usage restrictions and commercial Model-as-a-Service terms; it is not an unrestricted license.
Meta recommends deployment with use-case-specific system safeguards.

Publisher statements, not AI IQ measurements.

Sourcesresearch.meta.ai/blog/muse-spark-open huggingface.co/meta-models/Muse-Spark-Open-816B-A42B

About this score

How much of Muse Spark Open's IQ is measured, and how the rest is estimated.

15/ 39benchmark results

5/ 7dimensions measured

2estimated

0/5Abstract

2/5Math

5/6Academic

1/6Coding

0/5Computer Use

3/6Reliability

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 Muse Spark Open'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 Muse Spark Open

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

#1 overall

gpt-6.1-sol

IQ141 19 higherEffective cost$6.29

IQ122 same IQEffective cost$0.69

Most similar profile

grok-4.5

IQ122 same IQEffective cost$6.20

#2 overall

opus-5.5

IQ141 19 higherEffective cost$35.67

#3 overall

Gemini 4 Argon

IQ140 18 higher

Muse Spark OpenComparison model

Compare with any modelBrowse all comparisons →

Frequently asked questions

Quick answers about Muse Spark Open, 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 Muse Spark Open?: Muse Spark Open has an estimated IQ of 122, which ranks #37 out of 168 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 Muse Spark Open best at?: Among the dimensions with published benchmark results, Muse Spark Open ranks highest in Reliability (IQ 126, #20 of 168) and Academic Reasoning (IQ 130, #30 of 187). Its lowest-ranked measured dimension is Programmatic Reasoning (IQ 126, #42 of 169).
What is the EQ of Muse Spark Open?: Muse Spark Open 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 Muse Spark Open. AI IQ doesn't estimate EQ from other benchmarks, so the score appears once one of those results is published.
Which benchmarks has Muse Spark Open been tested on?: Muse Spark Open has published results on 15 of the 39 benchmarks behind AI IQ's Composite IQ, including SciCode, AA Omniscience, Humanity's Last Exam, MMMU-Pro and CritPt.
How much does Muse Spark Open cost?: AI IQ doesn't have a verified standard per-token API price for Muse Spark Open, so it can't calculate an effective cost. This is common for models offered only through subscriptions, restricted-access programs, or self-hosting.
Is Muse Spark Open open source?: Yes — Muse Spark Open is an open-weights model: its trained weights are published, so it can be downloaded and self-hosted.
Who developed Muse Spark Open?: Muse Spark Open was developed by Meta, based in the United States.
When was Muse Spark Open released?: Muse Spark Open was publicly released on October 15, 2026.
How large is the context window of Muse Spark Open?: Muse Spark Open supports a context window of 1,000,000 tokens — the amount of text it can consider in a single request.
How fast is Muse Spark Open?: In independent measurements, Muse Spark Open generates about 103 tokens per second, with a typical response time of about 25 seconds.