Muse Spark 1.2 (Minimal): benchmark results

Provider: Meta. Released 2026-08-05. Access: API.

Unified ELO 1695 ± 1, rank #153 of 3078 rated models, from 14 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
FlavourBench - Constraint54.99Constraint Family Score (0-100)96.2
Korean CSAT 2026 (Easy Mode) - Chemistry I50Points (out of 50)90.8
FlavourBench63.75FlavourBench Score (0-100)88.5
FlavourBench - Pairing70.33Pairing Family Score (0-100)76.9
Korean CSAT 2026 (Easy Mode) - Society and Culture43Points (out of 50)73.8
Context Arena62.83Average Score (%)67.2
Korean CSAT 2026 (Easy Mode) - Life Science I41Points (out of 50)67
Korean CSAT 2026 (Easy Mode) - Physics I38Points (out of 50)66.7
Korean CSAT 2026 (Easy Mode) - Total419.83Points (out of 450)66.7
Korean CSAT 2026 (Easy Mode) - English97Points (out of 100)66
FlavourBench - Substitution65.94Substitution Family Score (0-100)57.7
Korean CSAT 2026 (Easy Mode) - Korean93.5Points (out of 100)56.8

Interactive version: theaggregate.ai/model?slug=muse-spark-1-2-minimal · How It Works · Data refreshed daily, snapshot 2026-09-19.