← All models
Compare

DeepSeek-V3.2 vs Muse Spark

Add, remove, or swap models to compare them side by side.

DeepSeek-V3.2Muse Spark
AttributeDeepSeek-V3.2DeepSeek · 1 in the newsMuse SparkMeta AI · 2 in the news
Scores
Intelligence (ECI)146153
Coding4076
Math3464
Reasoning & Knowledge3482
Agentic & Tools3563
Specifications
DeveloperDeepSeekMeta AI
FamilyDeepSeek
ReleasedDec 1, 2025Apr 8, 2026
Parameters
AvailabilityOpen weights (unrestricted)API access
Context window164K
Price — $/M input$0.25
Price — $/M output$0.42
Inputstext
Outputstext
Benchmarks
AIME 2024/202588%89%
APEX7%
ARC-AGI57%
ARC-AGI-24%
FrontierMath22%39%
FrontierMath Tier 42%15%
GPQA Diamond83%90%
Humanity's Last Exam22%41%
LiveCodeBench86%
MMLU-Pro86%
SciCode39%52%
SimpleBench53%
SimpleQA Verified28%66%
Terminal-Bench40%
WebDev Arena12861513
WeirdML47%
τ²-bench91%92%

Use-case scores are 0–100 percentile composites across each area’s benchmarks, ranked against every model from the past year. Highlighted cells lead each row. Open a model for the full picture.

Benchmarks & model data from Epoch AI (CC BY); pricing & specs from OpenRouter. ECI = Epoch Capabilities Index.