Model comparison

Muse Spark vs Step 5 Preview

Muse Spark is the stronger model overall, scoring 50.6 to 47.9 on the Noometry Index.

Last verified . 17 shared benchmarks.

Muse Spark Meta

50.6

Rank #46 Confirmed

Step 5 Preview StepFun

47.9

Rank #58 Confirmed

Summary

  • They share 17 benchmarks with published results for both. Muse Spark scores higher in 5 categories and Step 5 Preview in 4 categories; 7 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where Muse Spark leads 65.7 to 41.2.
  • The biggest single-benchmark swing is ProofBench: 17% for Muse Spark and 42% for Step 5 Preview.

Side by side

Muse Spark and Step 5 Preview specifications
Muse SparkStep 5 Preview
ProviderMetaStepFun
Noometry Index50.647.9
Released2026-04-082026-09-16
WeightsProprietaryProprietary
Context window—1.02M
Max output—66K
Input $ / M tokens—$1
Output $ / M tokens—$2.70
Results tracked2718

Sponsored placements are available on pages like this one. Advertise on Noometry

Category by category

Coding Step 5 Preview leads

Muse Spark: 46.2 (#69), Step 5 Preview: 51.8 (#36)

Coding benchmarks
BenchmarkMuse SparkStep 5 Preview
SciCode51.5%58.9%
LMArena Coding14811480
LMArena WebDev—1564

Reasoning Step 5 Preview leads

Muse Spark: 35.9 (#67), Step 5 Preview: 40.0 (#57)

Reasoning benchmarks
BenchmarkMuse SparkStep 5 Preview
CritPt11.3%20.9%
LMArena Hard Prompts14741465
Epoch Capabilities Index152.04—

Math Muse Spark leads

Muse Spark: 47.8 (#66), Step 5 Preview: 46.1 (#72)

Math benchmarks
BenchmarkMuse SparkStep 5 Preview
ProofBench17%42%
LMArena Math14551470
OTIS Mock AIME 2024-202588.9%—
FrontierMath (Feb 2025 set)39%—
FrontierMath Tier 4 (v1)14.6%—

Knowledge Muse Spark leads

Muse Spark: 65.7 (#13), Step 5 Preview: 41.2 (#112)

Knowledge benchmarks
BenchmarkMuse SparkStep 5 Preview
LMArena Expert14571470
GPQA Diamond89.8%—
Humanity's Last Exam40.6%—

Multimodal Muse Spark leads

Muse Spark: 43.4 (#24), Step 5 Preview: 41.0 (#41)

Multimodal benchmarks
BenchmarkMuse SparkStep 5 Preview
LMArena Vision13061267
LMArena Document1444—

Multilingual Muse Spark leads

Muse Spark: 56.1 (#24), Step 5 Preview: 53.8 (#54)

Multilingual benchmarks
BenchmarkMuse SparkStep 5 Preview
LMArena Non-English14641432
LMArena Chinese15091519
LMArena Russian14661436
LMArena Spanish14721439
LMArena French1497—
LMArena German1497—
LMArena Korean1459—

Instruction Following Too close to call

Muse Spark: 75.9 (#51), Step 5 Preview: 76.0 (#50)

Instruction Following benchmarks
BenchmarkMuse SparkStep 5 Preview
LMArena Instruction Following14421444

Long Context Too close to call

Muse Spark: 44.4 (#69), Step 5 Preview: 44.5 (#64)

Long Context benchmarks
BenchmarkMuse SparkStep 5 Preview
LMArena Longer Query14511455

Writing & Preference Muse Spark leads

Muse Spark: 66.0 (#39), Step 5 Preview: 62.8 (#71)

Writing & Preference benchmarks
BenchmarkMuse SparkStep 5 Preview
LMArena Text14741442
LMArena Creative Writing14591410
LMArena Multi-Turn14771447

Frequently asked questions

Is Muse Spark better than Step 5 Preview?

Muse Spark is the stronger model overall, scoring 50.6 to 47.9 on the Noometry Index.

Is Muse Spark or Step 5 Preview better for coding?

Step 5 Preview scores higher on coding benchmarks: 51.8 versus 46.2 in the Noometry coding category.

How many benchmarks do Muse Spark and Step 5 Preview share?

17 benchmarks have published results for both models. Muse Spark has 27 scored results on Noometry and Step 5 Preview has 18.

Related comparisons

Go deeper