> Muse Spark is a closed AI model from Meta, released 8 Apr 2026. Index score 159 (GPT-4 = 100), rank 41 of 229 ranked models.

[Models](https://themodelindex.org/models/) /[Meta](https://themodelindex.org/labs/meta/)

# Muse Spark

Meta

Released 8 Apr 2026

Closed

Flagship

Date verified

[Compare](https://themodelindex.org/compare/muse-spark-vs-kimi-k2-6/) [Announcement](https://about.fb.com/news/2026/04/introducing-muse-spark-meta-superintelligence-labs/)

159148–171

Index score

#41 · provisional, limited data

2/ 8

Core tests taken

1 of 6 areas · plus 4 anchor tests

−11

Versus the frontier at release

best then: GPT-5.4

Replaced by [Muse Spark 1.1](https://themodelindex.org/models/muse-spark-1-1/)

–

Price per million tokens

no public API price found

–

Context window

tokens it reads at once

## Where it sits · on the frontier

_Chart: Model Index score of every scored model since late 2022, with the best closed and best open model over time_

Gemini 3 Pro reached this score 5 months earlier.

## Test results · by area · tick = best by any model

Show 4 anchor tests

Reasoning & knowledge

evidence

GPQA Diamond

90%

Humanity's Last Exam

41%

Maths

evidence

FrontierMath T1-3

no result

Coding

no evidence yet

SWE-bench Verified

no result

Terminal-Bench

no result

Agents

no evidence yet

APEX-Agents

no result

Novel problems

no evidence yet

ARC-AGI-2

no result

Long tasks

no evidence yet

METR time horizon

no result

Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.

## What others say · rank on each leaderboard

The Model Index

independent tests, our method

#41

of 219 · 159

LMArena · Text

overall chat quality, judged by people

#14

of 388 · 1489

LMArena · Creative writing

creative writing

#20

of 386 · 1465

LMArena · Vision

understanding images

#8

of 149 · 1294

LMArena · Coding

coding questions in chat

#21

of 383 · 1530

LMArena · Maths

maths questions in chat

#51

of 373 · 1466

Epoch Capabilities Index

overall capability from many benchmarks

#40

of 239 · 152.1

LMArena · Documents

working with long documents

#27

of 41 · 1444

Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source.

## Best at · top three among current models

2nd

understanding images

LMArena · Vision

Where this model places in the top three of today's models, on category leaderboards and on our core tests.

## Where to use it · prices checked 1 Oct 2026

#### From the lab

Meta AI

app · subscription

↗

Provider prices come from OpenRouter's live listing; going direct to a provider can differ. Subscriptions are the lab's own apps.

## Facts and sources

Released

8 Apr 2026

Announcement

[about.fb.com ↗](https://about.fb.com/news/2026/04/introducing-muse-spark-meta-superintelligence-labs/)

Release entry

Muse Spark (proprietary; MSL)

Scored as

Muse Spark

Artificial Analysis index

–

Output speed

–

Hugging Face

closed weights

## Meta releases · around this one

Llama 4 Maverick

Open

5 APR 2025

index 115

Llama 4 Scout

Open

5 APR 2025

index 111

Muse Spark

Closed

8 APR 2026

index 159

Muse Spark 1.1

Closed

9 JUL 2026

index 164

Muse Spark 1.2

Closed

5 AUG 2026

index 166

---
Source: https://themodelindex.org/models/muse-spark/ · The Model Index · data as of 2026-10-01