> Muse Spark 1.3 is a closed AI model from Meta, released 2 Sep 2026. Index score 170 (GPT-4 = 100), rank 17 of 229 ranked models. API price $1.25 in / $4.25 out per million tokens, 1.05M context.

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

# Muse Spark 1.3

Meta

Released 2 Sep 2026

Closed

Flagship

[Compare](https://themodelindex.org/compare/muse-spark-1-3-vs-kimi-k3/)

170159–182

Index score

#17 · provisional, limited data

2/ 8

Core tests taken

2 of 6 areas · plus 6 anchor tests

−22

Versus the frontier at release

best then: Claude Fable 5.1

$1.25/ $4.25

Price per million tokens

input / output, list price

1.05M

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_

GPT-5.4 reached this score 6 months earlier.

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

Show 6 anchor tests

Reasoning & knowledge

no evidence yet

GPQA Diamond

no result

Humanity's Last Exam

no result

Maths

evidence

FrontierMath T1-3

74%

Coding

no evidence yet

SWE-bench Verified

no result

Terminal-Bench

no result

Agents

evidence

APEX-Agents

58%

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

#17

of 219 · 170

LMArena · Maths

maths questions in chat

#7

of 373 · 1512

LMArena · Text

overall chat quality, judged by people

#8

of 388 · 1495

LMArena · Coding

coding questions in chat

#10

of 383 · 1539

Artificial Analysis Intelligence Index

overall capability across ten evaluations

#7

of 207 · 48

LMArena · Creative writing

creative writing

#25

of 386 · 1459

LMArena · Vision

understanding images

#10

of 149 · 1290

Epoch Capabilities Index

overall capability from many benchmarks

#17

of 239 · 156.9

LMArena · WebDev

building web apps and front ends

#12

of 123 · 1655

LiveBench · Overall

overall, with fresh questions to limit contamination

#6

of 58 · 81.6

LMArena · Documents

working with long documents

#14

of 41 · 1471

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

3rd

working with long documents

LMArena · Documents

3rd

agentic coding questions

LiveBench · Agentic Coding

3rd

instruction following questions

LiveBench · Instruction Following

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

↗

#### API providers · 1 · per million tokens, in / out

M

Meta

$1.25 / $4.25

cheapest

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

2 Sep 2026

Announcement

not linked yet

Release entry

Muse Spark 1.3

Scored as

Muse Spark 1.3

Artificial Analysis index

48 · max

Output speed

174 tokens/s

Hugging Face

closed weights

## Meta releases · around this one

Muse Spark 1.2

Closed

5 AUG 2026

index 166

Muse Glimmer 30B

Open

10 AUG 2026

Muse Spark 1.3

Closed

2 SEP 2026

index 170

---
Source: https://themodelindex.org/models/muse-spark-1-3/ · The Model Index · data as of 2026-10-01