Noometry

Model comparison

Muse Spark vs Qwen2.5 72B Instruct

Muse Spark is the stronger model overall, scoring 50.6 to 31.9 on the Noometry Index.

Last verified . 19 shared benchmarks.

Muse Spark Meta

50.6

Rank #46 Confirmed

Qwen2.5 72B Instruct Alibaba (Qwen)

31.9

Rank #267 Confirmed

Summary

  • They share 19 benchmarks with published results for both. Muse Spark scores higher in 8 categories and Qwen2.5 72B Instruct in 0 categories; 8 gaps are clear of the uncertainty.
  • The widest gap is in knowledge, where Muse Spark leads 65.7 to 27.0.
  • The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 88.9% for Muse Spark and 8.1% for Qwen2.5 72B Instruct.
  • Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.

Side by side

Muse SparkQwen2.5 72B Instruct
ProviderMetaAlibaba (Qwen)
Noometry Index50.631.9
Released2026-04-082024-09
WeightsProprietaryOpen
Context window—131K
Max output—8K
Input $ / M tokens—$1.40
Output $ / M tokens—$5.60
Results tracked2743

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

Category by category

Coding Muse Spark leads

Muse Spark: 46.2 (#69), Qwen2.5 72B Instruct: 33.2 (#260)

BenchmarkMuse SparkQwen2.5 72B Instruct
LMArena Coding14811292
SciCode51.5%—
WeirdML—16%
BigCodeBench Instruct—45.8%
BigCodeBench Complete—55.9%

Agentic & Tool Use Not comparable

Muse Spark: —, Qwen2.5 72B Instruct: 22.1 (#133)

BenchmarkMuse SparkQwen2.5 72B Instruct
TheAgentCompany—5.7%
BALROG—16.2%
METR Time Horizons—35.8%

Reasoning Muse Spark leads

Muse Spark: 35.9 (#67), Qwen2.5 72B Instruct: 22.3 (#199)

BenchmarkMuse SparkQwen2.5 72B Instruct
LMArena Hard Prompts14741271
Epoch Capabilities Index152.04129
CritPt11.3%—
DTBench—62.9%
LMCA—13.4%
BIG-Bench Hard—79.8%
ForecastBench—57.5
HellaSwag—84.8%
PIQA—82.6%
WinoGrande—82.3%

Math Muse Spark leads

Muse Spark: 47.8 (#66), Qwen2.5 72B Instruct: 19.3 (#287)

BenchmarkMuse SparkQwen2.5 72B Instruct
OTIS Mock AIME 2024-202588.9%8.1%
LMArena Math14551283
ProofBench17%—
Omni-MATH—33%
MATH Level 5—63.2%
FrontierMath (Feb 2025 set)39%—
FrontierMath Tier 4 (v1)14.6%—

Knowledge Muse Spark leads

Muse Spark: 65.7 (#13), Qwen2.5 72B Instruct: 27.0 (#253)

BenchmarkMuse SparkQwen2.5 72B Instruct
GPQA Diamond89.8%49.1%
LMArena Expert14571245
Humanity's Last Exam40.6%—
MMLU-Pro—63.1%
Confabulations—19.1%
GPQA (HELM)—42.6%
ARC (AI2) Challenge—94.5%
MMLU—85.3%
TriviaQA—71.9%

Multimodal Not comparable

Muse Spark: 43.4 (#24), Qwen2.5 72B Instruct: —

BenchmarkMuse SparkQwen2.5 72B Instruct
LMArena Vision1306—
LMArena Document1444—

Multilingual Muse Spark leads

Muse Spark: 56.1 (#24), Qwen2.5 72B Instruct: 41.0 (#213)

BenchmarkMuse SparkQwen2.5 72B Instruct
LMArena Non-English14641252
LMArena Chinese15091272
LMArena French14971280
LMArena German14971234
LMArena Korean14591188
LMArena Russian14661264
LMArena Spanish14721256
LMArena Japanese—1180

Instruction Following Muse Spark leads

Muse Spark: 75.9 (#51), Qwen2.5 72B Instruct: 65.5 (#221)

BenchmarkMuse SparkQwen2.5 72B Instruct
LMArena Instruction Following14421254
IFEval—80.6%

Long Context Muse Spark leads

Muse Spark: 44.4 (#69), Qwen2.5 72B Instruct: 38.9 (#188)

BenchmarkMuse SparkQwen2.5 72B Instruct
LMArena Longer Query14511282

Writing & Preference Muse Spark leads

Muse Spark: 66.0 (#39), Qwen2.5 72B Instruct: 46.7 (#215)

BenchmarkMuse SparkQwen2.5 72B Instruct
LMArena Text14741269
LMArena Creative Writing14591221
LMArena Multi-Turn14771272
WildBench—80.2%

Frequently asked questions

Is Muse Spark better than Qwen2.5 72B Instruct?

Muse Spark is the stronger model overall, scoring 50.6 to 31.9 on the Noometry Index.

Is Muse Spark or Qwen2.5 72B Instruct better for coding?

Muse Spark scores higher on coding benchmarks: 46.2 versus 33.2 in the Noometry coding category.

How many benchmarks do Muse Spark and Qwen2.5 72B Instruct share?

19 benchmarks have published results for both models. Muse Spark has 27 scored results on Noometry and Qwen2.5 72B Instruct has 43.

Related comparisons

Go deeper