Solar Pro 3: benchmark results

Upstage's enterprise reasoning MoE (102B total/12B active), Korean/English bilingual (January 2026). Provider: Upstage. Released 2026-01-27. Access: API.

Unified ELO 1547 ± 1, rank #453 of 1392 rated models, from 29 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
AA IFBench71.2Accuracy (%)86.9
AA TAU-2 Bench86.26Accuracy (%)78.9
AA Humanity's Last Exam10.29Accuracy (%)54.9
UGI - Natural Intelligence23.04NatInt Score54.5
AA GPQA Diamond72.42Accuracy (%)51.8
AA Omniscience - Health20.1Accuracy (%)48.1
AA Omniscience - Law11.2Accuracy (%)48.1
Artificial Analysis Intelligence Index8.53Intelligence Index44.6
BenchmarkList ECI105.52Capability Index (ECI)43.6
AA-Omniscience Accuracy18.5Accuracy (%)42.5
AA Omniscience - Software Engineering (SWE)22.5Accuracy (%)40.9
AA Terminal-Bench Hard7.58Accuracy (%)40

Interactive version: theaggregate.ai/model?slug=solar-pro-3 · How It Works · Data refreshed daily, snapshot 2026-09-05.